(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Children's Library | Biodiversity Heritage Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
Upload
See other formats

Full text of "Chemistry of cellulose and wood"

§> 



o 




; 




Florida Agricultural 
Experiment Station Library 




Gainesville, Florida 



iS^^Ks 





L 



Digitized by the Internet Archive 
in 2013 



http://archive.org/details/chemistryofcelluOOscho 

/ 



INTERNATIONAL CHEMICAL SERIES 
JAMES F. NORRIS, Ph.D., Consulting Editor 



THE CHEMISTRY OF 
CELLULOSE AND WOOD 



INTERNATIONAL CHEMICAL SERIES 

(James F. Norms, Ph.D., Consulting Editor) 



Adkins and McElvain — 

Elementary Organic Chemistry 
Practice of Organic Chemistry 

Bancroft — 

Applied Colloid Chemistry 

Bingham — 

Fluidity and Plasticity 

Bogue — 

The Theory and Application 
of Colloidal Behavior 

Cody- 
General Chemistry 
Inorganic Chemistry 

Daniels — 

Mathematical Preparation for 
Physical Chemistry 

Daniels, Mathews and Williams — 
Experimental Physical Chemis- 
try 
Eucken, Jette and LaMer — 

Fundamentals of Physical 
Chemistry 
Griffin — 

Technical Methods of Analysis 
As Employed in the Labora- 
tories of Arthur D. Little, Inc. 
Hall and Williams — 

Chemical and Metallographic 
Examination of Iron, Steel 
and Brass 
Hamilton and Simpson — 

Calculations of Quantitative 
Chemical Analysis 
Hammett — 

Solutions of Electrolytes 
Leighou — 

Chemistry of Engineering 
Materials 
Loeb — 

Proteins and the Theory of 
Colloidal Behavior 
Long and Anderson — 

Chemical Calculations 
Lord and Demorest — 

Metallurgical Analysis 
Mahin — 

Quantitative Analysis 
Introduction to Quantitative 
Analysis 
Mahin and Carr — 

Quantitative Agricultural 

Analysis 
Mellon — 

Chemical Publications 



Millard — 

Physical Chemistry for Colleges 

Moore — 

History of Chemistry 
Norris — 

Experimental Organic Chem- 
istry 

Inorganic Chemistry for 
Colleges 

Organic Chemistry 

Norris and Mark — 

Laboratory Exercises in Inor- 
ganic Chemistry 

Parr — 

Analysis of Fuel, Gas, Water 
and Lubricants 
Reedy — 

Elementary Qualitative Analy- 
sis for College Students 

Rice— 

Organic Chemistry 

Robinson — 

The Elements of Fractional 
Distillation 

Schorger — 

The Chemistry of Cellulose 
and Wood 

Smith and Miller — 

An Introduction to Qualitative 
Chemical Analysis and the 
Related Chemical Principles 

Stone and Dunn — 

Experiments in General Chem- 
istry 

Underwood — 
Problems in Organic Chemistry 



The Colloidal Salts 
The Hydrous Oxides 

White — 

Technical Gas and Fuel Analy- 
sis 
Wilkinson — 

Calculations in Quantitative 
Chemical Analysis 
Williams and Homerberg — 

Principles of Metallography 
Woodman — 

Food Analysis 



The late Dr. H. P. Talbot was Consulting Editor of the 
International Chemical Series from its inception in 1911 until 
his death in 1927. 




THE CHEMISTRl! OM 
CELLULOSE AND f mf&U 



6RGIA 



i 

) 



BY 

A. W. SCHORGER, PH. D. 

Director of Chemical Research, C. F. Burgess Laboratories, 

Inc. Madison, Wis. (Formerly Research Chemist, Forest 

Products Laboratory, U. S. Forest Service) 



Go 

o 

CD 



First Edition 
Second Impression 



CO 






McGRAW-HILL BOOK COMPANY, Inc. 
NEW YORK: 370 SEVENTH AVENUE 

LONDON: 6 & 8 BOUVERIE ST., E. C. 4 

1926 



S 3 7 * c 



Copyright, 1926, by the 
McGraw-Hill Book Company, Inc. 



PRINTED IN THE UNITED STATES OF AMERICA 



THE MAPLE PRESS COMPANY, YORK, PA. 



Dedicated 
To My Associates 

in THE 

C. F. BURGESS LABORATORIES, Inc. 






v'sai 



PREFACE 

It is the aim of this book to cover completely the scientific 
and empirical data available on the chemistry of wood. This 
has necessitated the inclusion of cotton cellulose and its modified 
forms. Industrial processes are treated mainly from the side 
of the fundamental reactions involved. Perhaps owing to the 
insufficiency of our knowledge, these reactions receive but scant 
attention in most technical books. A thorough understanding 
of them, however, would remove pulping processes, e.g., from the 
category of an art to that of a science. Many of the problems 
connected with the utilization of cellulose and wood cannot be 
solved by chemistry alone, so that there have been included 
certain chapters from the domain of biological chemistry and 
colloid science. 

It is hoped that the usefulness of the book will be increased by 
the detailed statement of the conditions under which certain 
results were obtained. Mere statement of fact would, in the 
majority of cases, be meaningless. This policy has been pursued, 
also, for the reason that much of the original literature is difficult 
of access. 

Cellulose, in its various forms, is the most abundant organic 
compound in the plant world and has been used by man since time 
immemorial. For a century or more, it has inspired cycles of 
chemical research, of varying intensity, the most pronounced 
being during the World War. At present, investigation is being 
pursued by so many workers, and from so many different angles, 
that it is to be hoped that interest will not flag until the important 
structural relationships are established. Lack of knowledge of 
the constitution of cellulose and lignin need occasion neither 
surprise nor impatience. The vast number of organic compounds 
has been largely produced from simple chemicals by synthesis. 
The research was prosecuted on the basis of chance and prob- 
ability and was not hampered by knowledge of an existing sample 
which had to be duplicated or torn apart. Progress was from the 



Vlll PREFACE 

simple to the complex, while with cellulose and lignin the reverse 
must be true. 

The chemistry of cellulose and wood, of lignin particularly, is 
surrounded by a mass of data, largely speculative, which is much 
in need of weeding. But the information is so indefinite that he 
who performs the operation at this stage, may later awake to the 
embarassing fact that the grain rather than the tares have been 
uprooted. Categorical statements, however desirable, have 
accordingly been largely avoided. Reaumur could call a crocodile 
un furieux insecte to his own satisfaction, but not to that of the 
present generation. 

A. W. Schorger 

Madison, Wisconsin 
June, 1926 



CONTENTS 

Page 

Preface vii 

CHAPTER I 

The Structure, Formation, and Physical Properties of Wood 1 

Wood Elements 1 

Formation of Wood 3 

Lignification 7 

Growth and Structure of the Cell Wall 9 

Physical Effects of Water 11 

Density . 18 

Refractive Index 20 

Double Refraction 20 

Fluorescense 21 

Specific Heat 21 

Thermal Conductivity 22 

CHAPTER II 

The Composition of Wood. . 24 

History 24 

Constituents of Wood 31 

Elementary Composition of Wood 31 

Analytical Composition of Wood 33 

Effect of Storage on the Composition of Wood 39 

Relation between Density of Wood and Its Cellulose Content . . 39 

Heartwood and Sapwood 41 

Ash 50 

Pectin 54 

Starch 59 

Sugars 62 

Nitrogen 63 

Secondary Constituents 64 

Fat, Resin, and Wax 65 

Cutin 68 

Irritant Constituents 68 

CHAPTER III 

Lignin 70 

Lignin Isolated with Sulphite Solutions 72 

ix 



X CONTENTS 

Page 

Cross and Bevan's Formula for Lignin 83 

Lignin Isolated with Alkalis 86 

Lignin Isolated with Strong Mineral Acids 91 

Condensation Products of Lignin 98 

Action of Halogens 101 

Acetyl and Formyl Groups 104 

Methoxyl Groups 108 

Oxidation of Lignin 114 

Nitration of Lignin 117 

Distillation of Lignin 118 

CHAPTER IV 

Color Reactions of Wood 122 

Color Reactions of Wood with Inorganic Compounds ...... 122 

Color Reactions of Wood with Organic Compounds 126 

CHAPTER V 

Hemicelluloses and Wood Cellulose 141 

Furfuroids 149 

Methylpentosans 152 

Preparation of Hemicelluloses 153 

Hydrolysis of Hemicelluloses 155 

Wood Gum. Xylan 157 

Esters and Ethers 161 

Araban 162 

Laevulan 162 

Mannan 163 

Galactan 166 

Wood Cellulose 167 

CHAPTER VI 

The Constitution op Cellulose 175 

Acetolysis of Cellulose 175 

Cellobiose 178 

Procellose 180 

Constitution of Cellobiose and Cellulose 180 

Optical Rotation .' 194 

CHAPTER VII 

Gelatinized Cellulose 196 

Coloration with Iodine 197 

Hygroscopic and Colloid Water 198 

Theories of Gelatinization 201 

Dyeing Properites 206 

Gelatinization with Salts 207 



CONTENTS xi 

Page 

Adsorption of Salts 209 

Gelatinization with Alkalis. Mercerization 213 

Alkali-cellulose. Adsorption of Alkalis 217 

Tests for Mercerization 226 

Gelatinization with Acids 228 

Adsorption of Acids 229 

Gelatinization by Beating 231 

Action of Bacteria 237 

Gelatinization of Lignocellulose 238 

Cellulose Solvents. Cuprammonium Solution 240 

Crystallization of Cellulose 247 

Salts as Cellulose Solvents 248 

Hydrogen Peroxide as a Cellulose Solvent 252 

Acids as Cellulose Solvents 252 

Viscose 254 

Laboratory Preparation of Viscose 257 

CHAPTER VIII 

OXYCELLULOSE 262 

Formation with Nitric Acid 262 

Oxycellulose by Chloric Acid 263 

Oxycellulose by Chromic Acid 264 

Oxycellulose by Permanganates 265 

Oxycellulose by Hypochlorites 266 

Oxycellulose by Salts and Air 267 

Oxycellulose by Alkalis and Air 268 

Oxycellulose by Hydrogen Peroxide 268 

Oxycellulose from Cuprammonium Solution 270 

Oxycellulose by Reduction of Cellulose Nitrate 270 

Oxycellulose by Electrolysis 271 

Cellulose Peroxide 272 

Properties of Oxycellulose 274 

Effect of Heat 275 

Esters of Oxycellulose 275 

Detection of Oxycellulose 276 

Formation of Furfural 279 

a-, /$-, and 7-Oxycelluloses and Their Salts 281 

Composition of Oxycellulose 283 

Nature of Oxycellulose 283 

CHAPTER IX 

The Action of Acids on Cellulose 290 

Hydrocellulose 290 

Organic Acids 292 

Salts 293 



Xii CONTENTS 

Page 

Preparation 294 

Properties of Hydrocellulose 297 

Effect of Heat 298 

Action of Acids and Alkalis 299 

Solubility 301 

Action of Dyes 302 

Hydrocellulose Nitrate 303 

Formula 303 

Nature of Hydrocellulose . . . '. -304 

Amyloid and Parchment 305 

Guignet's Soluble Cellulose 308 

Mechsig's Amyloid 308 

Parchment 308 

Cellulose Dextrins and Sulphuric Esters 310 

Cellulose Acetates 316 

Cellulose Nitrates 320 

Nitrates from Wood Pulp 326 

Purification of Wood Pulp 329 

Hydro xymethy If urfural 331 

CHAPTER X 

Saccharification of Cellulose and Wood 335 

Saccharification of Cellulose with Concentrated Acids 335 

Saccharification of Cellulose with Dilute Acids 341 

Sugars and Ethyl Alcohol from Wood 345 

Saccharification with Dilute Acids 345 

Source of Ethyl Alcohol 352 

Saccharification with Concentrated Acids 354 

Products 357 

CHAPTER XI 

The Action of Various Reagents on Wood 359 

Action of Heated Water 359 

Action of Cellulose Solvents 363 

Action of Bases 365 

Formation of Oxalic Acid 366 

Action of Acids 369 

Action of Salts 372 

Fire Retardants 373 

Resistance of Wood to Chemicals 375 

Weathering and Natural Coloration of Wood 379 

Artificial Coloration of Wood 381 

Dyeing Wood 382 

Action of Light 383 

Action of Oxidizing Agents 384 



CONTENTS xm 

CHAPTER XII 

Page 

Pulp Processes and Wood Pulps 387 

Sulphite Process 387 

By-products 398 

Soda Process 403 

By-products . 406 

Sulphate Process 408 

Sodium Sulphite Process 412 

Minor Pulping Processes 413 

Wood Pulps 415 

Discoloration of Sulphite Pulp 423 

Bleaching 424 

CHAPTER XIII 

The Distillation of Cellulose and Wood 428 

Destructive Distillation of Cellulose 428 

Vacuum Distillation of Cellulose 432 

Destructive Distillation of Wood 434 

Exothermic Reaction 436 

Yields of Products 441 

Distillation with Superheated Steam 445 

Effect of Moisture 445 

Order of Formation of Products 446 

Pressure above Atmospheric 447 

Temperature Control 447 

Use of Catalyzers 450 

Charcoal 453 

Constitution and Composition of Charcoal 458 

Composition of Distillates 459 

Source of Products 463 

CHAPTER XIV 

The Fermentation of Cellulose and Wood by Bacteria and Fila- 
mentous Fungi 470 

Fermentation of Cellulose by Bacteria 470 

Fermentation of Cellulose by Filamentous Fungi 481 

Fermentation of Wood by Filamentous Fungi 482 

Products of Decay 494 

CHAPTER XV 

Digestion of Cellulose and Wood by Animals 496 

CHAPTER XVI 

Analytical Methods 505 

Preparation of Sample 505 



xiv CONTENTS 

Page 

Determination of Water 505 

Cold-water-soluble 506 

Hot-water-soluble 506 

Alkali-soluble 506 

Ether and Alcohol Extracts 506 

Ash 507 

Determination of Nitrogen 507 

Determination of Cellulose 508 

Chlorine Dioxide Method 516 

Preparation of Chlorine Dioxide Solution 516 

Titration of Chlorine Dioxide Solution 516 

Determination of Lignin 518 

Hydrochloric Acid Method 519 

Sulphuric Acid Method 519 

Determination of Lignin in Pulps 526 

Determination of Pectin 526 

Determination of Methyl Alcohol 527 

Determination of Methoxyl 529 

Determination of Acetyl Groups in Wood 530 

Determination of Acetyl Groups in Cellulose Acetate 531 

Determination of Pentosans 532 

Phloroglucinol Method 534 

Bromine Method 536 

Determination of Mannan 537 

Determination of Galactan 538 

Determination of a-Cellulose 539 

Baryta Resistance Value 541 

Copper Number 541 

Determination of Oxycellulose 544 

Preparation of Standard Cellulose 545 

Determination of the Bleach Requirement of Pulps 547 

Tingle Bro mate Method .547 

Roe Chlorine Method 549 

Differentiating Unbleached Sulphite from Sulphate Pulp 552 

Differentiating Sulphite from Soda Pulp 552 

Differentiating Bleached from Unbleached Sulphite Pulp . . . 554 

Herzberg Iodine Stain 554 

Lignin Reagents 555 

Index 557 



CHEMISTRY OF CELLULOSE 
AND WOOD 



CHAPTER I 

THE STRUCTURE, FORMATION, AND PHYSICAL 
PROPERTIES OF WOOD 

In many trees the most prominent feature of the cross-section 
of the trunk is the separation into dark heartwood (duramen) and 
light sapwood (alburnum). In certain species, such as spruce, 
cottonwood, aspen, beech, and hackberry, there is little difference 
between the colors of heartwood and sapwood. Closer inspection 
shows that the wood, if grown in a temperate climate, consists of 
annual rings which are usually quite sharply defined. The 
inner portion of the ring, which is light in color, represents the 
growth produced in spring, and the dark outer portion that pro- 
duced in summer. The dark summer wood is always more 
dense than the spring wood. 

Wood is composed of hollow cells of various kinds, collectively 
known as wood elements. The cavity is known as the lumen. 
All cells, without doubt, perform a useful service, but in some 
cases their function is not clear. The minute structures of the 
two main classes of woods, gymnosperms (softwoods) and 
dicotyledons (hardwoods) , differ greatly from each other. 

Wood Elements. — Most of the cells in wood are long and 
narrow. Wood fibers (Fig. le), which are largely limited to hard- 
woods give strength and stiffness. Tracheids predominate in the 
conifers, where they perform the dual function of affording 
mechanical strength and conducting sap. Tracheids are polygo- 
nal in cross-section and arranged in radial rows. A character- 
istic feature of these elements is the presence of bordered pits. 
Holes or thin spots in the wall are known as pits; in shape these 

1 



CHEMISTRY OF CELLULOSE AND WOOD 



vary from slits to elliptical and round forms (Figs. Id and 2DE). 
Elements have simple pits when their walls are parallel with the 
primary layer or middle lamella and bordered pits (Fig. 2B) 
when they diverge from it. The bordered pits are usually 

arranged in exactly similar position to 
pits in adjoining elements. The open- 
ings are closed by partitions which are 
extensions of the middle lamella. The 
partitions usually contain a thickened 
disk or torus. The portion of the mem- 
brane surrounding the torus is fre- 
quently perforated, as is proved by the 
fact that it is possible to force India 
ink through it. 1 

The hardwoods contain cells of rela- 
tively large diameter superimposed to 
form a continuous tube, called a pore 
or vessel (Fig. lc). These vessels serve 
to conduct sap from the roots to the 
leaves. The conifers do not have ves- 
sels, but the sap travels from one tra- 
cheid to another through the membrane 
of the pit. When the pores are of 
approximately the same size, as in 
maple and birch, the woods are known 
as diffuse porous, and when there are 
concentric rows of large and small 
pores, as in oak and elm, as ring porous. 
The vessels of dicotyledons are fre- 
ce£?^7o""o" q«en% closed by tyloses. These are 
parenchyma; b, two cells from pith-like cells which are formed by 

^of^eLltf oSs iDt ™ si0n int0 the lumen 0f the WSSel 

x leading into its upper and of wood or ray parenchyma. Tyloses 
\"^ T %%X%? ; are rarely found in conifers except in 

the resin canals. 

Medullary or pith ray cells (Fig. la) run in a procumbent 
position from the pith. They serve to conduct sap from the 
bark to the sapwood and vice versa, and to store starch. 

1 1. W. Bailey, Forestry Quart., 11 (1913), 12-20. 




PHYSICAL PROPERTIES OF WOOD 3 

Parenchyma are short, thin-walled cells which occur in the 
xylem as scattered vertical rows, of which the upper and lower 
cells are tapered. These cells contain protoplasm and retain 
their power of subdivision after leaving the cambium. Their 
function is the same as that of the ray parenchyma cells. 




Fig. 2. — (A) Piece of radial wall of a tracheid showing a bordered pit in surface 
view. Part of the dark-colored torus or thickened portion of the pit membrane 
may be seen through the opening in the wall. This pit orifice is surrounded or 
bordered by the embossed portion of the secondary wall. (B) Section cut 
through (A) at a-b, showing embossed wall seen in (A) and that of cell directly 
under it. The membrane and torus occupy a median position. (C) Section 
showing the valve-like action of the torus. (Br) Bordered area, (ME) membrane, 
(Ts) torus. (D) Tracheid from spring-wood. (E) Tracheid from summer wood. 
The tracheids have their pitted radial walls uppermost. (After Bailey.) 



Formation of Wood. — The structural membranes of the higher 
plants are formed from the carbon dioxide of the air by the aid 
of light. The first acceptable working hypothesis of photosyn- 
thesis was that of Baeyer. 1 He assumed that when sunlight 
strikes chlorophyll which is surrounded by carbon dioxide the 
latter is decomposed into oxygen, which is liberated, and carbon 
monoxide, which remains combined with the chlorophyll. The 

1 A. Baeyer, Ber., 3 (1870), 68. 



CHEMISTRY OF CELLULOSE AND WOOD 



carbon monoxide was reduced to formaldehyde, which 
polymerized to sugar. 

CO + H 2 = H.CHO. 
6H.CHO = C 6 H 12 6 . 

A step in advance was the opinion of Hoppe-Seyler 1 that the 
carbon dioxide actually combined as such with the chlorophyll 
to form a loose compound. Willstatter and Stoll 2 have con- 
firmed this view. The chlorophyll-formaldehyde-peroxide 



:N 



:N 



X co 2 

"^Mg > 

/ H 2 



^N 



.- Mg-O-C > 

\)H 



/ 



^NH 



^N 



^N 



^r Mg-o-c 



Chlorophyll Carbon dioxide compound Chlorophyll- 

of chlorophyll formaldehyde-peroxide 

breaks up into chlorophyll, formaldehyde, and oxygen. It is 
probable that the formaldehyde polymerizes immediately. 

A very large literature has accumulated on the photosynthesis 
of carbohydrates. There is no satisfactory evidence that in the 
living plant the process passes through the formaldehyde stage, 
though the detection of formaldehyde has been reported. Baly 3 
and his associates found that when an aqueous solution of carbon 
dioxide was exposed to ultra-violet light of wave length 200^ 
formaldehyde was produced; this, in turn, was polymerized to 
reducing sugars at wave length 290/xju. Both processes occur 
simultaneously in the light of a quartz mercury lamp, so that 
formaldehyde cannot be detected except by removing it with a 
stream of carbon dioxide or preferably by the addition of sub- 
stances which absorb the polymerizing light of wave length 
290jU)U. The reaction is reversible in the unscreened light of a 

*F. Hoppe-Seyler, " Physiologische Chemie," Berlin (1881), p. 139. 

2 R. Willstatter and A. Stoll, " Untersuchungen iiber die Assimilation 
der Kohlensaure," Berlin (1918), p. 244. 

3 E. C. C. Baly, I. M. Heilbron, and W. F. Barker, /. Chem. Soc, 
119 (1921), 1025-1035; J. Soc. Chem. Ind., 40 (1921), 377-379R. 



PHYSICAL PROPERTIES OF WOOD 5 

quartz mercury lamp and the equilibrium lies far towards the 
side of carbon dioxide. In the presence of a photocatalyst such 
as chlorophyll the equilibrium is reversed and thrown to the side 
of the reducing sugar. This explains the difficulty encountered 
in detecting formaldehyde as an intermediate product. 

The syrup obtained by the polymerization of formaldehyde by 
ultra-violet light 1 contained about 10 per cent of a hexose sugar. 2 
Polysaccharides and anhydro-sugars were absent. 

The method whereby the cell walls are built up from the simple 
products of photosynthesis in the leaf is far from clear. The 
secondary wood, which constitutes almost all of the wood of the 
stem, is produced in the cambium layer, which is situated between 
the cortex and the last layer of sap wood. The cambium cells 
are soft and thin walled; each contains protoplasm, a nucleus, 
and organic and inorganic constitutents essential to growth. The 
first stage in the multiplication of the cell consists in the splitting 
of the nucleus and the formation of protoplasmic spindle fibers 
between the nuclei. These enlarge at the middle to form a cell 
plate, which splits to produce the plasma membrane of the two 
daughter cells. Between these membranes is secreted a sub- 
stance indistinguishable from lignin in the mature cell, which 
forms the primary layer or middle lamella. Upon the middle 
lamella are deposited layers of cellulose intermingled with lignin. 

One of the cells forms bark or wood, while the other is left to 
undergo further subdvision. An annual ring in the radial 
direction may represent as many as 100 wood elements. The 
cells reach full development rapidly and then die. 

Wislicenus 3 has advanced the theory that growth of the cell 
wall takes place by adsorption of colloids from the sap of the 
cambium. Cellulose is first formed as a chemically indifferent 
framework possessing a large surface, which is thickened by 
adsorption of a mixture of colloids to produce lignin. Experi- 
mental work was performed with sap obtained by tapping. 
Before the period of foilage development the amount of colloids 

1 E. C. C. Baly, J. Ind. Eng. Chem., 16 (1924), 1016-1018. 

2 J. C. Irvine and G. V. Francis, J. Ind. Eng. Chem., 16 (1924), 1019- 
1020. 

3 H. Wislicenus and M. Kleinstuck, Kolloid-Z., 6 (1910), 17-23, 87-94; 
H. Wislicenus, Ibid., 27 (1920), 209-223; Cellulosechemie, 6 (1925), 45-58. 



6 



CHEMISTRY OF CELLULOSE AND WOOD 



in the sap is small and increases slightly during the period of 
wood formation. Only 3.5 to 8.4 per cent of the weight of the 
total solids in the spring sap of birch consisted of colloids. This 
was determined by adsorbing the colloids by shaking with 
alumina, filtering, and determining the solids in the nitrate. 
Of the total solids in the sap of sugar maple only 4.4 per cent 
were colloids, while that of hornbeam contained 21.08 per cent 
before the period of leaf formation. The fresh sap gave no 
reaction for starch with iodine solution and no reaction for lignin. 

The sap of the birch contained the following total solids and 
colloids : 



Date 



Solids per 

100 cubic 

centimeters, 

grams 



Amount 
adsorbed, 
per cent 



March 12 
April 15. . 
April 23 . . 
May 26.. 
July 27... 



0.6070 
0.6964 
1.2238 
0.4154 
0.2292 



4.45 

6.70 

8.30 

37.07 

19.55 



It is not at all apparent how the theory of Wislicenus can be 
applied to the growth of the cell wall. The amount of colloids 
in the sap is low and it is unlikely that they can penetrate, in quan- 
tity, either the middle lamella or the plasma membrane in which 
the vital syntheses take place. All the evidence points to the 
presence of protoplasm at every point where growth takes place; 
hence it is most probable that transformation of sugars into col- 
loids and formation of the cell wall take place simultaneously at 
the middle lamella-plasma membrane interface. Casparis 1 con- 
siders it improbable that incrustation with lignin takes place 
through adsorption of external material by the cell membrane. 
It is more likely that lignin is formed in situ from carbohydrates. 

It is desirable to follow the migration of the carbohydrates 
from the leaf. The starch formed in the leaf by photosynthesis 
is converted into sucrose, ostensibly by enzymes, and transported 
downward to the growing cells through the sieve tubes of the 

1 P. Casparis, C. A., 15 (1921), 1333. 



PHYSICAL PROPERTIES OF WOOD 



cortex. Neither the flow of water upward in the sapwood or of 
sap downward in the cortex has been satisfactorily explained. 
Diffusion can play but a minor part in the transportation of 
foodstuffs. It would require 2 years and 7 months for 1 milli- 
gram of sucrose to travel from a 10 per cent solution a distance of 
1 meter. 1 Starch as such is not transportable. The sugar, 
after reaching the ray cells and wood parenchyma, may be 
reconverted into starch and held in reserve. 

An investigation by Jones 2 on the storage of sugar in the maple 
is particularly illuminating as to the probable method of growth 
of the cell wall. Maple sap, as it flows from the tree, is rich in 
sucrose but practically free from reducing sugars; if the wood is 
ground and extracted with water, approximately one-half of the 
sugar obtained consists of reducing sugars. When water is 
forced through a section of wood, sucrose can be washed out, but 
not reducing sugars. This shows conclusively that sucrose is 
stored in the pores and reducing sugars in the cell sap of the living 
cells. It would, accordingly, appear that the synthesis of constit- 
uents of the cell wall begins with carbohydrates no higher than 
monoses. 

Lignification. — A study of the lignin content of rye straw 
shows that a large amount of lignin is present in the early stages 
of growth. 3 Rye stalks 10 days after planting in summer con- 
tained 11.6 per cent of lignin and this value increased to 14.5 
per cent in 8 weeks. The change in lignin and methoxyl in the 
stem of rye sown in October is shown below. It would appear 
that cellulose and lignin are formed simultaneously; furthermore, 
since lignification increases with age it is probably not due to 
adsorption of colloids from without, as the cell wall would be 
expected to decrease in permeability as it approached maturity. 



Age, in days 



192 



223 



231 



239 



253 



274 



Lignin (Willstatter), per cent. 
CH 3 in lignin, per cent 



13.03 
3.03 



17.24 
11.46 



18.57 
13.41 



18.86 
12.84 



19.07 
13.38 



20.49 
13.42 



1 H. de Vries, Botan. Z., 43 (1885), 1. 

2 C. H. Jones, J. Assoc. Official Agr. Chem., 2 (1916), 103-111. 

3 E. Beckmann, O, Liesche, and F. Lehmann, Biochem. Z., 139 (1923), 
502. 



8 CHEMISTRY OF CELLULOSE AND WOOD 

Little reliance can be placed on staining with zinc chloriodine 
as a measure of lignification. Potter 1 found that in numerous 
woods, sections of the cell wall gave the blue color of cellulose. 
This was particularly true after boiling the sections in water; the 
wood of JEsculus after boiling in water showed bluish-violet 
layers in nearly all the fibers and cells. Spaulding 2 found that 
when the wood of Populus was stained immediately after felling 
the thick inner layer of the cell was invariably stained blue. 

The function of lignin is generally assumed to be physical. 
According to Sachs, 3 lignification produces an increase in the 
hardness of the cell membrane and a decrease in extensibility; it 
permits the membrane to be easily penetrated by water without 
swelling appreciably. Sonntag 4 found that the ability of fibers 
to swell in water, particularly in cross-section, decreased with the 
degree of lignification. Tracheids of Abies pectinata increased 
14.8 per cent in cross-section from the air-dry condition, while 
fibers of pure linen increased 55.5 per cent. Lignification pro- 
duces decrease in ductility and tensile strength. 

Lignification, aside from increasing the hardness of the cells, 
renders them more resistant to crushing and bending. 5 

Schellenberg, 6 on the other hand, has held that a membrane is 
not changed in its mechanical properties by lignification, but 
shows the same gradations in the magnitude of its strength, 
extensibility, and ability to swell as unlignified fibers. The 
physical significance of lignification lies in the fact that a ligni- 
fied membrane is no longer capable of growth. Lignification 
always takes place at a time when the cell contains a living 
plasma. The cell wall having become lignified, it no longer shows 
surface growth and apparently no increase in thickness; the cell 
is no longer capable of division. 

It is now known that the amount of lignin present in a wood has 
no direct relation to its mechanical properties. In general, the 

1 M. C. Potter, Ann. Botany, 18 (1904), 121-140. 

2 P. Spaulding, Rept. Missouri Botan. Gardens, 17 (1906), 58. 

3 J. Sachs, "Lehrbuch der Botanik," 4 Aufl., Leipzig (1874), p. 21. 

4 P. Sonntag, Landw. Jahrb., 21 (1892), 839-869. 

s F. Lukas, Akad. Wiss. Wien, 85, I (1882), 292-327; 87, I (1883), 303- 
327; cf. T. von Weinzierl, Ibid., 76, I (1877), 385-461. 

6 H. Schellenberg, Jahrb. wiss. Botanik, 29 (1896), 237-266. 



PHYSICAL PROPERTIES OF WOOD 



9 



best single criterion for strength is specific gravity. That a 
wood, such as bald cypress, having a high lignin content, need 
not necessarily have high mechanical properties is shown in the 
following table: 



Lignin, 1 
per cent 



Specific 
gravity 



Modulus 

of 

rupture, 2 

pounds 

per square 

inch 



Modulus 
of 

elasticity, 

pounds 

per square 

inch 



Hardness, 
oak =100 



Bald cypress . . 
Yellow poplar 
Yellow birch . . 



33.9 
23.2 

26.3 



0.45 
0.37 
0.55 



7,110 
5,570 

8,810 



1,378 
1,207 
1,490 



34 
32 
70 



1 G. J. Ritter and L. C. Fleck, J. Ind. Eng. Chew,., 15 (1923), 1055. 

2 H. S. Betts, "Timber," N. Y. (1919). 



Growth and Structure of Cell Wall.— In 1853, von Mohl 1 
expressed the opinion that the wall grows by apposition, i.e., the 
deposition of substances in successive layers. Nageli, as a 
result of his study of the growth of starch grains 2 and the cell 
wall, 3 concluded that the cell wall consists of ultramicroscopic, 
crystalline, molecular complexes which he called micellae. By 
this assumption he was able to explain striation, stratification, 
swelling, double refraction, and other properties of the cell wall. 

Each micella is normally surrounded by a film of water, but in 
the dry condition the micellae form a continuous membrane. 
Swelling takes place up to the point where the pressure of the 
aqueous layers equals the cohesion between the micellae. The 
smaller the micellae the greater is their attraction for water. 
The attraction of the micellae for water decreases relatively more 
rapidly than that between micellae. The water closest to the 
micellae is most firmly held, while this attraction decreases at a 
distance to such an extent that some of the water is merely held 
by capillarity in the spaces created by swelling. The spiral 
arrangement of the micellae explains why shrinking and swelling 
occur laterally and not longitudinally. 



1 H. von Mohl, Botan. Z., 11 (1853), 753, 769; 16 (1858), 373. 

2 C. W. Nageli, "Die Starkekorner," Zurich (1858), 623 pp. 

3 C. W. Nageli, Beitr. Wiss. Botan., 3 (1863), 1-126; C. W. Nageli and 
S. Schwendener, "Das Mikroskop," 2 Aufl., Leipzig (1877), 679 pp. 



10 CHEMISTRY OF CELLULOSE AND WOOD 

Nageli 1 assumed that the cell wall grew by intussusception, 
that is, the new micellae were formed between the old. The old 
micellae gradually increased in size, whereby their ability to hold 
water gradually decreased. Growth by intussusception has been 
disputed in numerous papers, but there is now general agreement 
that it usually takes place in this way, though occasionally by 
apposition, or both. Nageli's general conception of the structure 
of the cell wall must be viewed as a great intellectual achieve- 
ment and as the most satisfactory yet proposed. 

The micellar theory has been applied to dyeing. 2 The 
penetration of dyes and mordants into the interstices between 
the micellae is limited to particles of a diameter less than 1/z/*. 
When the fibers have been mercerized larger particles can be 
absorbed, since the size of the interstices is increased. Rontgen- 
ray examination in the wet and dry state of normal cellulose, 
mercerized fibers, and threads produced from viscose and a 
cuprammonium solution of cellulose shows no definite change in 
the lattice of the crystals as a result of swelling. 3 The change 
produced is accordingly intermicellar. 

Another theory of the structure of the cell wall is of passing 
interest. Wiesner 4 thought that the cell wall was formed of 
round microscopic bodies which he called dermatosomes. These 
were obtained under oxidizing and hydrolytic conditions, whereby 
the fibers were converted into brittle oxycellulose or hydrocellu- 
lose. So far as known, these fragments have no bearing on the 
structure of the wall. 

In recent years there has been much ingenious though fanciful 
speculation on rhythmic growth. Liesegang 5 found that, when 
a drop of concentrated silver nitrate solution was placed on a 
gelatin plate containing potassium bichromate, formation of 
silver chromate did not take place uniformly, but in the 
shape of concentric rings. No adequate explanation has so far 
been offered for the phenomenon. Kuster 6 has attempted to 

1 C. W. Nageli, I.e.; "Ueber das Wachsthum der Starkekorner durch 
Intussusception," Botan. Mitt., 3 (1881), 487-534. 

2 R. Haller, Kolloid-Z., 20 (1917), 127-145. 

3 J. R. Katz, Z. Physik, 25 (1924), 321-326. 

4 J. Wiesner, Akad. Wiss. Wien, 93, I (1886), 17-80. 

5 R. Liesegang, "Chemische Reaktionen in Gallerten," Leipzig (1898). 

6 E. Kuster, "Ueber Zonenbildung in kolloidalen Medien," Jena (1913). 



PHYSICAL PROPERTIES OF WOOD 11 

explain the growth of wood as due to banding in a colloidal 
medium. When the end of a cylinder of gelatin was immersed in 
silver nitrate he obtained not only rings but ascending spiral 
bands of silver chromate. In this connection Balls 1 has noted 
that when a cotton hair is swelled to five to ten times its normal 
thickness with sodium hydroxide and carbon bisulphide, as many 
as 25 daily growth rings become visible under the microscope. 
The thickness of each ring is about 0.4/*. 

Physical Effects of Water. — Water occurs in wood as "free" 
water in the cell cavities and as " imbibed" water in the cell 
walls. Some of the water must be colloidally bound with the 
fibers. The natural state is a condition wherein the cell walls 

Fiber Saturation Points Determined by Compression Tests 1 
(Results based on the dry weight of the wood) 

Moisture 

(Average), 

Species Per Cent 

Longleaf pine, green 25 

Red spruce, green 31 

Chestnut, green 25 

Loblolly pine: 

Heartwood, green 23 

Heartwood, air dry 24 

Sapwood, green 24 

Sapwood, air dry 26 

Norway pine: 

Heartwood, green 30 

Sapwood, green 28 

Tamarack, green 30 

Western hemlock, green 29 

Red fir, green 23 

White ash, green 20 

Red gum, air dry 25 

Red spruce : 

Air dry 30-32 

Superheated 24-25 

Tamarack : 

Air dry 30-32 

Superheated 24-25 

Chestnut : 

Air dry 26-27 

Superheated 22-24 

'H. D. Tiemann, Forest Service Circular, 108 (1907); "The Kiln Drying of Lumber," N. Y. 
(1917), p. 104. 

1 W. L. Balls, Proc. Roy. Soc., 90B (1919), 542-555. 



12 



CHEMISTRY OF CELLULOSE AND WOOD 



are saturated with water; this is known as the fiber saturation 
point. There is no sharply defined method for determining this 
point. The methods usually employed are based on the observa- 
tion that as soon as the water falls below the fiber saturation 
point the wood begins to shrink and increase in strength. The 
logical procedure would be to expose the wood to an atmosphere 
of 100 per cent humidity until the cell walls are saturated; how- 
ever, the method is complicated by the condensation of water on 
the surface of the wood, so that free water is present. The fiber 
saturation point varies from 20 to 30 per cent of the dry weight 
of the wood. The results obtained by compression tests are 
shown in the accompanying table. 

The influence of the density of wood and its resin content on 
hygroscopicity has been investigated by Zeller. 1 In all cases 
where the specific gravities lay between 0.41 and 0.80, and the 
resin contents were small, the curves for hygroscopic moisture 
were the same up to a certain point from which they diverged 
according to the density of the wood. At a relative humdity 
of 94.75 to 96 per cent, the less dense woods took up the greatest 
amount of moisture. This is explained on the assumption that 

Moisture in Shortleaf Pine Sapwood Shavings at Various Humidities 

and at 25°C. 



Sp. gr. 0.41 


Sp. gr. 0.61 Sp. gr. 0.69 


Moisture 

in 

shavings, 

per cent 


Relative 

humidity, 

per cent 


Moisture 

in 
shavings, 
per cent 


Relative 

humidity, 

per cent 


Moisture 

in 
shavings, 
per cent 


Relative 

humidity, 

per cent 


4.12 
7.80 
9.00 
10.80 
15.50 
18.75 
23.80 
27.60 
30.25 
34.00 


11.0 
33.2 
49.5 
60.4 
81.0 
90.0 
95.5 
97.5 
98.7 
100.0 


2.5 
7.9 
9.2 
10.4 
16.2 
20.0 
24 . 25 
26.00 
27.60 
29.88 


6.6 
35.8 

47.5 
61.3 
82.2 
90.0 
96.0 
97.4 
98.7 
100.0 


3.0 
7.88 
9.7 
10.75 
15.3 
19.3 
23.4 
25.0 
26.25 
28.00 


7.5 
34.6 
53.0 
62.5 
79.3 
91.0 
95.3 
97.5 
98.8 
100.0 



S. M. Zeller, Ann. Missouri Botan. Gardens, 7 (1920), 51-74. 



PHYSICAL PROPERTIES OF WOOD 



13 



at and above this point moisture is condensed and absorbed on 
the surface of the fibers. The point of divergence of the curves 
accordingly represents the fiber saturation point. The fiber 
saturation points for longleaf and shortleaf pine determined in this 
way were: 



Heartwood 
moisture, 
per cent 



Sapwood 
moisture, 
per cent 



Longleaf pine . 
Shortleaf pine . 



23.25 
24.5 



23.75 
24.25 



Resin has little influence on hygroscopic moisture until the 
humidity approaches 100 per cent. At 100 per cent humidity, 
heartwood of longleaf pine of sp. gr. 0.51 showed a hygroscopicity 
of 33.2 per cent, while wood of sp. gr. 0.86 to 0.94 with 18.4 
per cent resin showed a hygroscopicity of 21.12 per cent. 

Beadle 1 appears to have first observed that the exposure of 
dry cellulose to water or its vapor produced a pronounced rise in 
temperature. The more finely divided the cellulose the more 
rapidly it took up moisture, but the coarse cellulose eventually 
took up more moisture than the fine. Two grams of cotton wool 
dried at 220°F. showed a rise of 4.8°F. after an exposure of 15 
minutes to moist air. The temperature rose more rapidly 
when air was blown against the cotton. He attributed the 
phenomenon to " hydration," since the cotton remained at a 
temperature greater than its surroundings as long as it was 
taking up moisture. 

Evolution of Heat with Starch and Water 

Water to 100 Grams Heat per Gram 

of Dry Starch, of Dry Starch, 

Grams Calories 

0.23 28.11 

2.39 22.60 

3.23 20.97 

8.16 12.43 

12.97 7.37 

15.68 5.21 

19.52 2.91 

'C. Beadle, Nature, 49 (1894), 457; Chem. News, 71 (1895), 1-2; C. 
Beadle and O. W. Dahl, Ibid., 73 (1896), 180-183. 



14 CHEMISTRY OF CELLULOSE AND WOOD 

Rodewald 1 has determined the heat evolved when starch 
containing various amounts of water swells in water. 

Masson 2 explained the rise in temperature as due to the 
formation of a solid solution, or what Travers 3 prefers to call a 
" rigid" solution. Chemical combination and condensation 
on the surface of the fiber are not directly involved. The rise 
and fall of the temperature curves are similar in water vapor 
and water ; in both cases the rise in temperature is due to conden- 
sation on the cotton of water vapor which is absorbed. The 
water probably undergoes osmotic diffusion into the fiber to 
form a solid solution whose vapor pressure is always lower than 
that of water. This is in accordance with Trouton's 4 observation 
that cotton containing a definite proportion of water behaves 
like an aqueous solution in that at different temperatures it 
shows a vapor tension which is a constant fraction of that of 
pure water. When cotton is plunged into water the air adhering 
to the fibers maintains the space necessary for the occurrence of 
distillation of the water, the vapors then condensing on the cotton. 

A rise in temperature is not peculiar to cellulose. In 1822, 
Pouillet 5 observed that when a powder was moistened with a 
liquid in which it did not dissolve there was a rise in temperature. 
Dry charcoal and silica will develop about 18 calories per gram 
when treated with water. 6 Parks 7 states that the heat evolved 
is proportional to the area of the surface exposed by the solid, 
and that the amount of heat developed per square centimeter is 
approximately 0.00105 calorie when the temperature is near 7°. 
According to Martini, 8 the surface exposure does not always 
explain the marked rise in temperature. He compares the 
Pouillet effect to a kind of " inverted solution." 

Dunlap 9 found that dry sugar maple, beech, and longleaf pine 
liberated 16.6 to 19.6 calories per gram when in contact with 

1 H. Rodewald, Z. physik. Chem., 24 (1897), 206. 

2 O. Masson, Proc. Roy. Soc, 74 (1905), 230-255. 

( 3 M. W. Travers, Proc. Roy. Soc, 79A (1907), 204-205. 

4 F. T. Trouton, Proc. Roy. Soc. 77A (1906), 292. 

5 Pouillet, Ann. chim. phys., [2] 20 (1822), 141-162. 

6 T. Martini, Phil. Mag., [5] 50 (1900), 618-619. 

7 G. J. Parks, Phil. Mag., [6] 4 (1902), 247. 

8 T. Martini, Phil. Mag., [6] 5 (1903), 595-596. 

9 F. Dunlap, Unpublished Report, Forest Products Laboratory. 



PHYSICAL PROPERTIES OF WOOD 



15 



water. In drying wood it is, accordingly, necessary to furnish 
this amount of heat to overcome the attraction of the wood for 
moisture. 

The hygroscopicity of absorbent cotton has been determined 
by Masson and Richards 1 by exposing it over sulphuric acid of 
constant vapor pressure. The cotton was brought to constant 
weight by exposure to phosphorus pentoxide for 24 hours. The 
results are given in the table below. The hygroscopic moisture 
present in cotton at a humidity of 97.2 per cent was 17.70 



Hygroscopicity < 


3F Cotton over Sulphuric Acid Solutions 


H2SO4 
Per cent 


V 
P 


Ma 


Me 


M 


M 

W 


73.8 


0.050 


0.0120 


0.0145 


0.0132 


0.0139 


67.0 


0.100 


0.0175 


0.0198 


0.0186 


0.0196 


59.0 


0.198 


0.0264 


0.0311 


0.0288 


0.0304 


52.6 


0.294 


. 0356 


0.0406 


0.0381 


. 0402 


47.2 


0.408 


0.0441 


0.0497 


0.0469 


0.0495 


43.1 


0.500 


0.0509 


0.0593 


0.0551 


0.0581 


40.5 


0.556 


. 0530 


. 0655 


0.0592 


. 0624 


38.8 


0.598 


0.0599 


0.0690 


0.0644 


0.0679 


32.3 


0.710 


0.0716 


0.0840 


0.0778 


0.0821 


26.8 


0.794 


0.0860 


0.1002 


0.0931 


0.0982 


23.2 


0.844 


0.0989 


0.1107 


0.1048 


0.1106 


20.3 


0.874 


0.1045 


0.1250 


0.1148 


0.1210 


18.1 


0.894 


0.1114 


0.1300 


0.1207 


0.1274 


10.3 


0.952 


0.1378 


0.1606 


0.1492 


0.1574 


6.2 


0.972 


0.1563 


0.1792 


0.1678 


0.1770 



W = weight (0.948 gram) of dry cotton. 
T = temperature (20°) of the apparatus. 

p = actual pressure of water vapor in the atmosphere employed. 
P = saturation pressure of water vapor at the same temperature. 
Ma = weight of moisture absorbed by the cotton after exposure in the 

apparatus till further absorption appears negligible. 
Me = weight of moisture retained by the sample after it has been super- 
saturated by exposure over water and then allowed to evaporate 
in the apparatus till further loss appears negligible. 
M = arithmetic mean of M A and Me] taken as indicating the amount 
of absorbed moisture which is required to establish true equilibrium. 
The per cent of absorbed water is obtained by multiplying the last 
column by 100. 

1 O. Masson and E. S. Richards, Proc. Roy. Soc, 78A (1906), 412-429. 



16 CHEMISTRY OF CELLULOSE AND WOOD 

per cent. According to the authors, the true absorption value 
can be obtained only by taking the mean of the apparent equilib- 
rium values reached by absorption and evaporation. Either 
change can be followed by observing the characteristic temper- 
ature as given by a thermometer. According to Rakovski, 1 
cellulose loses water more slowly than it absorbs it. The 
maximum hysteresis is 1.8 per cent of water. 

Fenchel and Comely 2 working with paper and paper 
pulps found that moisture was absorbed more rapidly than it 
was lost. The initially rapid absorption of moisture in a 
saturated atmosphere decreased greatly in rate and was not 
complete in 14 days; after 17 days at a temperature of 18° 
(± 0.5°), cotton absorbed 16.5 per cent of water and spruce 
mechanical pulp, 22.35 per cent. According to Scheurer, 3 dry, 
bleached cotton will take up 19.0 to 20.2 per cent of water from 
an atmosphere saturated with water vapor. 

Wood shows considerable variation in hygroscopicity between 
species, cotton having an intermediate value. 4 

Browne 5 found that cellulose absorbed only 12.57 per cent of 
water after exposure at 20° for 25 days in an atmosphere of 100 per 
cent humidity. This datum is abnormal. 

Absolutely dry cellulose is stated to be the most hygroscopic 
substance known. It will take up 3 per cent of moisture from 
all known drying agents. 6 

The amount of water absorbed is influenced by the temperature 
and humidity of the surrounding atmosphere, and various other 
factors. The water-soluble constituents of cotton took up 28.71 
per cent of moisture in comparison with 8 per cent for cotton 
under the same conditions. Heating 7 cellulose above 100° 



1 A. V. Rakovski, J. Russ. Phys.-Chem. Soc, 47 (1915), 18-21; cf. A. K. 
Urquhart and A. M. Williams, J. Textile Inst., 15 (1924), 138; T. Kujira 
et al., J. Soc. Chem. Ind., 43 (1923), 894A. 

2 K. Fenchel and R. Cornely, Wochbl. Papierfabr., 44 (1913), 4323-4327. 

3 A. Scheurer, Bull. Mulhouse, 87 (1921), 129-135. 

4 H. D. Tiemann, "The Kiln Drying of Lumber," 3rd ed. (1920), p. 105. 

5 C. A. Browne, «/. Ind. Eng. Chem., 14 (1922), 712; cf. R. E. Wilson 
and T. Fuwa, Ibid., p. 913. 

6 H. Jentgen, Z. angew. Chem., 23 (1910), 1544. 

7 J. H. Lester, J. Soc. Chem. Ind., 21 (1902), 388-389. 



PHYSICAL PROPERTIES OF WOOD 17 

reduces the hygroscopicity. 1 Cotton cloth which has been 
thoroughly dried will not regain the amount of moisture held in 
the air-dry state even after long exposure to the atmosphere. 2 
Complete drying by the aid of heat appears to produce a state of 
partial irreversibility, so that hygroscopicity should always be 
determined with material dried at low temperatures. The hygro- 
scopicity of wood is reduced by heating in air or steam at 
temperatures above 100°. 3 

The absorption of moisture has been explained as a case of 
solid solution. It is a general law in chemistry that like dissolves 
like. Cellulose with its hydroxyl groups shows a pronounced 
affinity for substances, such as water and alcohol, which contain a 
hydroxyl group. Stradnikov 4 has shown that water requires 
several months to displace alcohol from cellulose at the ordinary 
temperature. Shrinkage of wood may be considerably reduced 
by the Powell process, wherein the wood is impregnated with a 
solution of molasses or sucrose, dried, and heated at 120° to 
produce caramelization. Concentrated solutions of sucrose con- 
tain large colloidal aggregates and are adhesive until crystalliza- 
tion sets in. 5 Crystalline sugar can penetrate the cell wall and 
subsequent caramelization would change it to a state of rigid, 
colloidal dispersion. Reduction in shrinkage may be due to 
mechanical resistance offered by colloidal sugar, but it is more 
probable that there is formation of a solid solution whereby the 
sugar partially satisfies the attraction of the cellulose for water. 

Wood in swelling exerts a pressure sufficient to crush the fibers 
transversely. This pressure in even the weakest wood is over 200 
pounds per square inch. The Egyptians are stated to have split 
stone by drilling holes and inserting dry wooden wedges which 
were then moistened with water. 6 The pressures exerted by 
some colloids is enormous. According to Rodewald, 7 1 gram of 

1 W. Will, J. Soc. Chem. Ind., 34 (1905), 148. 

2 S. H. Higgins, J. Soc. Chem. Ind., 28 (1909), 188; J. Hubner and F. 
Teltscher, Ibid., 28 (1909), 644. 

3 H. D. Tiemann, "The Kiln Drying of Lumber," (1917), p. 260. 

4 G. L. Stradnikov, /. Russ. Phys.-Chem. Soc, 48 (1916), 301. 

5 H. Wislicenus, Cellulosechemie, 6 (1925), 57. 

6 G. Rawlinson, " History of Ancient Egypt." 

7 H. Rodewald, Landw. Vers.-Sta., 45 (1895), 227. 



18 CHEMISTRY OF CELLULOSE AND WOOD 

starch can develop a pressure of 2523 atmospheres on swelling 
in water. 

The radial shrinkage of most woods varies from 2 to 8.5 per 
cent, and the tangential from 4.2 to 14 per cent. The lower 
shrinkage radially is due to the arrangement of the medullary 
ray cells at right angles to the axis of the tree. The longitudinal 
shrinkage is very slight; this holds for masses as well as for indi- 
vidual fibers. Koehler 1 states that theoretically the cell walls 
are made of fibrils which run spirally in the cell walls. The 
water is held almost entirely between these fibrils. The fibrils 
are ultramicroscopic, but their probable orientation can be 
determined by the direction of checks and other markings in the 
cell wall. These spirals usually run nearly parallel to the axis of 
the cell; as the water leaves the spaces between the fibrils they 
draw together in a direction almost at right angle to the axis. 
This action results in a considerable transverse shrinkage, but in 
slight longitudinal shrinkage. His theory of the structure of the 
cell wall and its behavior with water closely resembles the 
micellar theory of Nageli; the latter was able to render visible 
the spiral striations in cells of Abies excelsa by swelling them in 
concentrated sulphuric acid. 2 

The manner in which moisture travels through wood is 
unknown. It appears to pass in the form of a vapor from one 
cell wall to the other, where it condenses and the process is 
repeated. Beams 12 inches square and 16 feet long require about 
2 years of air drying before the moisture in the interior falls to 
the fiber saturation point. 3 

Density. — Dunlap 4 has found the average density of wood 
substance to be 1.54. There are appreciable differences between 
species as shown in the following table : 

Longleaf pine 1 .5060 Beech 1 .4990 

Douglas fir 1.5639 Red oak 1.5395 

Pacific yew 1 .5534 Sugar maple 1 .5506 

1 A. Koehler, "The Properties and Uses of Wood," N. Y. (1924), p. 50. 

2 C. W. Nageli and S. Schwendener, "Das Mikroskop" 2 Aufl., Leipzig 
(1877), p. 536. 

3 H. D. Tiemann, Forest Service Bull, 70 (1906), 123. 

4 F. Dunlap, J. Agr. Research, 3 (1914), 423-428. 



PHYSICAL PROPERTIES OF WOOD 



19 



The results were obtained by determining the density of the 
solution of calcium nitrate in which the wood would remain in 
suspension. The method is obviously inaccurate, since the 
fibers would not absorb the calcium nitrate solution as such but 
water preferentially. 

Richter 1 determined the density of wood and wood pulps in 
alcohol in a pycnometer. The density of the pulps varied 
from 1.408 to 1.634. Pure cellulose had a density of 1.583. 
The values for three species of American woods were : 

Black spruce (Picea americana) 1 .512 

White spruce (Picea canadensis) 1.517 

Balsam, heartwood (Abies fraseri) 1 . 402 

De Mosenthal 2 gives the density of wood cellulose as 1.575 
and of cotton as 1.61. Apparently, the density of the encrusting 
substances is less than that of cellulose. The great variations in 
densities reported for both wood and cellulose indicate the 
necessity for further investigation. 

The specific gravity of some common American woods in the 
oven-dry state are given below: 

Apparent Specific Gravities of Some American Woods 





Specific 
gravity 




Specific 
gravity 


Ash, white 


0.60 
0.42 
0.40 
0.66 
0.66 
0.46 
0.54 
0.66 
0.71 


Cedar, white 

Cypress, bald 

Fir, Douglas 

Hemlock 

Pine, jack 


0.32 


Aspen 


0.47 


Basswood 


0.52 


Beech 


0.44 


Birch, yellow 


0.46 


Chestnut 


Pine, Norway 

Pine, longleaf 


0.51 


Elm, white 


0.64 


Maple, sugar 


Spruce, white ... 


0.43 


Oak, white 


Tamarack 


0.56 









Woods of the highest specific gravity are found in tropical and 
subtropical regions. The black ironwood {Rhamnidium ferreum) 
of Florida has the highest specific gravity, namely, 1.302, of any 
wood native to the United States. 

1 E. Richter, Wochbl. Papierfabr., 46 (1915), 1529-1533. 

2 H. de Mosenthal, J. Soc, Chem, Ind., 26 (1907), 445. * 



20 CHEMISTRY OF CELLULOSE AND WOOD 

From 40 to 80 per cent of the total volume of wood is occupied 
by air. Since the density of wood substance is approximately 
1.54, it may be calculated that 100 parts of wood having a spe- 
cific gravity of 0.40 can take up 185 parts of water. Maumene 1 
tested 32 species of woods and found that the highest absorption 
was shown by chestnut, 100 parts of which took up 174.86 parts of 
water. 

Refractive Index. — Celluloses obtained by denitration of the 
nitrates of wood cellulose, ramie, flax, and cotton had about the 
same refractive index, namely n D = 1.5310. 2 

Double Refraction. — According to Tunmann, 3 lignified 
membranes are more strongly refractive than pure cellulose 
membranes. Under polarized light, dense walls show a dark 
cross. On ultramicroscopic examination, they show an approxi- 
mately parallel series of large, strongly luminous micellae between 
which are optically empty spaces. Remec 4 states that lignin has 
no influence on the specific double refraction of plant fibers. 

It has been assumed until recently that the double refraction 
of cellulose was due to the presence of anisitropic crystals. 
According to Harrison, 5 double refraction is due to internal 
stresses running parallel to the axis of the fiber. It disappears 
when the strain is relieved by swelling the fiber in cuprammo- 
nium solution except in those bands unaffected by the reagent. 

It has been found by Ambronn 6 that when fine strands 
of substances are imbedded parallel to each other so as to form 
a grating in a medium having a different refraction, the system 
behaves as a uniaxial positive crystal. If the medium has a 
similar refractive index, this effect disappears while true double 
refraction persists. Under the latter conditions, cellulose fibers 
obtained by the denitration of collodion show a strong double 
refraction which clearly indicates crystalline anisitropy. Katz 7 
states that the double refraction of cellulose is most simply 

1 E. J. Maumene, Compt. rend., 87 (1878), 943-946. 

2 H. de Mosenthal, /. Soc. Chem. Ind., 26 (1907), 447. 

3 O. Tunmann, "Pflanzenmikrochemie," (1913), p. 593. 

4 E. Remec, Akad. Wiss. Wien, 110, I (1901), 387. 

5 W. Harrison, Proc. Roy. Soc, 94A (1908), 460-469. 

6 H. Ambronn, Kolloid-Z., 18 (1916), 90-97, 273-281. 

7 J. R. Katz, Kolloidchem. Beihefte, 9 (1917), 160. 



PHYSICAL PROPERTIES OF WOOD 21 

explained on the assumption that the fibers are built up of micellae 
or minute crystals. 

Fluorescence. — In 1893, Hartley 1 found that blotting paper 
was strongy fluorescent and capable of rendering visible the 
whole of the ultra-violet spectrum as far as wave length 2000. 
This fluorescence lacks distinct color. 

The subject has been extensively investigated by Lewis. 2 
Mechanical pulp is devoid of fluorescent properties and cellulose 
nitrate nearly so. Normal cellulose from all sources gives a 
nearly uniform spectrum. Mercerized cellulose, viscose, parch- 
ment, and paper from well-beaten pulps show a characteristic 
and distinct effect at wave length 2750. Each modified cellulose 
has a characteristic curve. It appears that constituent groups 
within the cellulose molecule affect the fluorescent properties, 
as indicated by distortion of the curve, so that it may eventually 
be possible to detect differences in constitution in this way. 

Specific Heat. — Fleury 3 obtained 0.366 for the specific heat of 
dry cellulose and 0.41 for cellulose containing 7 per cent of water. 
The following values were obtained by Dietz : 4 

Specific 
Heat 

Artificial silk 0.324 

Filter paper 0.319 

Sulphite pulp . 319 

Soda pulp 0.323 

Straw pulp . 325 

Mechanical pulp, pine . 327 

Cotton 0.319 

There evidently is little difference between the specific heats of 
cellulose and lignocellulose. 

Dunlap 6 determined the specific heats of a large number of 
woods. The samples were dried at 105°. There was little 
variation between species. The portion of the tree from which 

1 W. N. Hartley, J. Chem. Soc, 63 (1893), 253. 

2 S. J. Lewis, J. Soc. Dyers Colourists, 34 (1918), 167-172; 37 (1921), 
201-204; 38 (1922), 68-76, 99-108. 

3 G. Fleury, Compt. rend., 130 (1900), 437. 

4 O. Dietz, Wochbl. Papierfabr., 43 (1912), 3119-3123. 

5 F. Dunlap, "The Specific Heat of Wood," Forest Service Bull, 110 
(1912), 28 pp. 



22 



CHEMISTRY OF CELLULOSE AND WOOD 



the samples were taken and the locality in which the trees grew 
were without effect. The mean value was 0.327. The specific 
heat is greatly influenced by temperature and is represented by 
the equation 

Specific heat = 0.266 + 0.001 16*, 

in which 0.266 represents the specific heat at 0°. Padoa 1 gives 
the specific heat of cellulose as 0.347 between and 80°. 

Thermal Conductivity. — The thermal conductivity of cellulose 
and wood fibers is much greater than that of air, so that the 
greater the density of the wood the poorer is its insulating prop- 
erty. The following values were obtained by van Dusen : 2 



Material 



Remarks 



Thermal 
conductivity 



KX10 



Density 



D 



Balsa wood. . . . 

Ceiba wood 

Balsa wood .... 

Sawdust 

Planer shavings 
Balsa wood. . . . 

Cypress 

White pine. . . . 

Mahogany 

Virginia pine . . . 

Oak 

Maple 



Very light; 

across grain 
Across grain 
Across grain 
Various 
Various 
Heavy 
Across grain 
Across grain 
Across grain 
Across grain 
Across grain 
Across grain 



10.7 

11.3 

11.9 
14.0 
14.0 
20.0 
23.0 
27.0 
31.0 
33.0 
35.0 
38.0 



7.5 

7.9 
8.3 
9.7 
10.0 
14.0 
16.0 
19.0 
22.0 
23.0 
24.0 
27.0 



0.113 

0.113 

0.118 

0.190 

0.140 

0.33 

0.46 

0.50 

0.55 

0.55 

0.61 

0.71 



7.1 

7.1 

7.4 
12.0 

8.8 
20.0 
29.0 
32.0 
34.0 
34.0 
38.0 
44.0 



thermal conductivity in cal. sec. x cm. l deg. 



k = thermal conductivity in B.t.u. per 24 hours, per square foot, 

per degree Fahrenheit, per inch of thickness. 
k = 69,700 K. 

D = density in grams per cubic centimeter. 
d = density in pounds per cubic foot. 
d = 62.5Z). 

1 M. Padoa, Atti accad. IAncei, rend., 29 (1920), 201. 

2 M. S. van Dusen, Trans. Am. Soc. Heating, Ventilating Eng., 26 (1920), 
406; cf. C. Niven and A. E. M. Geddes, Proc. Roy. Soc, 87A (1912), 535- 
539; T. S. Taylor, Mech. Eng., 42 (1920), 8-10. 



PHYSICAL PROPERTIES OF WOOD 23 

The conductivities of the woods measured fall roughly on a 
straight line when plotted against density. 

Railway ties 6 by 8 inches in cross-section required steaming 
at 20 pounds pressure for an average period of 4 hours and 20 
minutes before the center reached a temperature of 100°. 1 

1 G. M. Hunt, Proc. Am. Wood-Preservers' Assoc., 11 (1915), 85-99. 



CHAPTER II 
THE COMPOSITION OF WOOD 

The development of our knowledge of the composition of plant 
tissues and of the fundamentals of present analytical methods is 
due preeminently to the earlier French investigators. 

History. — Payen 1 observed that the carbon content of oak, 
beech, and poplar was reduced by treatment with alkali. The 
"thickening substance" deposited in the lignified cells differed 
from the membrane proper and was attacked by alkalis and nitric 
acid. Rasped wood was treated with 20 times its weight of 
concentrated nitric acid, allowed to stand 30 hours, and washed 
with sodium hydroxide. There remained a residue of cellulose, 
isomeric with starch. Analysis showed : 2 

Per Cent 

C 43.85 

H 5.86 

50.28 

Wood was recognized as consisting of two principal substances, 
cellulose and "true ligneous matter." The " matter e incrustante" 
had a higher carbon content 3 than the cellulose, analysis showing : 

Per Cent 

C 53.76 

H 6.00 

O 40.20 

The existence of incrusting material was denied by Fremy, 4 who 
thought that the cell walls contained several fiber substances 
differing from one another in their solubility in various 
chemicals. The starch-free cells of roots and fruits were treated 
with cuprammonium solution to dissolve the cellulose. The 

1 A. Payen, Compt. rend., 7 (1838), 1052-1056. 

2 A. Payen, Compt. rend., 7 (1838), 1125; 8 (1839), 52. 

3 A. Payen, Compt. rend., 8 (1839), 169. 

4 E. Fremy, Compt. rend., 48 (1859), 202-208. 

24 



THE COMPOSITION OF WOOD 25 

green-colored inner membrane remaining consisted of a salt of 
pectic acid, which was made soluble by treatment with hydro- 
chloric acid followed by caustic potash. Fungi and the pith of 
certain trees behaved differently under the same treatment, so 
that he assumed the existence of various celluloses. This, in 
turn, was denied by Payen, 1 who attributed the variable proper- 
ties of cellulose from different sources to the difference in degree 
of adhesion between the cellulose and the compounds with which 
it was associated. 

It was then shown by Fremy 2 that density played but a small 
part with respect to solubility in cuprammonium solution, since 
vegetable ivory dissolved rapidly while wood fibers did not. As 
proof of the existence of definite isomeric substances, he showed 
that rice paper, insoluble in its original condition, became soluble 
in cuprammonium solution after treatment with dilute mineral 
acids or caustic alkalis. In this way cellulose could be extracted 
from paper but not from fungi. Payen 3 maintained his original 
position that the cellulose in woody plants did not differ essen- 
tially from that of cotton and other textile fibers ; when sufficiently 
free from incrusting materials, all dissolved in Schweizer's reagent. 

Pelouze 4 showed that cellulose precipitated from cupram- 
monium solution was then soluble in hydrochloric acid of a 
concentration in which ordinary cellulose would not dissolve; 
that concentrated hydrochloric acid would dissolve cellulose, 
water precipitating it in a form like that from a cuprammonium 
solution. Payen 5 later recognized the existence of isomeric 
celluloses, but maintained that it had not yet been shown that the 
soluble cellulose could be distinguished chemically from that in 
plants; the latter could be rendered soluble in cuprammonium 
solution, by treating the tissues with glacial acetic acid, without 
destroying the structure of the cells. 

In a paper devoted specifically to the chemical composition of 
wood, Fremy 6 rejected the idea that wood consists of cellulose 

1 A. Payen, Compt. rend., 48 (1859), 210-221. 

2 E. Fremy, Compt. rend., 48 (1859), 275-279. 

3 A. Payen, Compt. rend., 48 (1859), 319-325. 

4 J. Pelouze, Compt. rend., 48 (1859), 210, 327. 

5 A. Payen, Compt. rend., 48 (1859), 328, 358. 

6 E. Fremy, Compt. rend., 48 (1859), 862-868. 



26 CHEMISTRY OF CELLULOSE AND WOOD 



impregnated with foreign substances. The term "cellulose" 
was reserved for substances like cotton and vegetable ivory 
which were directly soluble in cuprammonium solution. Wood 
fibers were first treated with dilute potassium hydroxide to 
remove tannins, albumin, and pectic substances, followed by 
hydrochloric acid of varying strength, fuming acid being 
finally used; the fibers were then soluble in cuprammonium 
solution. He finally treated them with concentrated sulphuric 
acid to remove the " utricular or fibrous" parts unaffected by the 
previous treatments. In this way were isolated vasculose, 
paracellulose, and fibrose. 

Vasculose was the substance of which the vessels and tracheids 
were composed. It was insoluble in acids and cuprammonium 
solution but dissolved in boiling concentrated potassium 
hydroxide. The medullary rays contained paracellulose; it was 
insoluble in cuprammonium solution but became soluble after 
treatment with acids and alkalis, or under the influence of heat 
and moisture. Pieces of oak and fir after treatment with con- 
centrated potassium hydroxide gave a white fibrous residue 
like paper pulp, called fibrose. His description of the latter is 
somewhat ambiguous. It was insoluble in cuprammonium 
solution in which cellulose dissolves at once and in the alkaline 
solution which dissolved the medullary rays. Fibrose was to be 
distinguished from cellulose in that when water was added to its 
solution in concentrated suphuric acid a thick jelly precipitated; 
cellulose under the same conditions gave dextrins not precipitable 
by water. 

In a later paper the opinions of Fremy 1 as to the composition 
of wood were modified to accord more closely with those of Payen. 
Analysis of wood comprised determination of " ligneous cuticle" 
or "cuticular substance," the "incrusting substance" of Payen, 
and true cellulose. The " ligneous cuticle" represented the 
residue remaining after treatment with sulphuric acid (H 2 S0 4 + 
2H 2 0), and washing with water and alkali. Cellulose was deter- 
mined by suspending the wood in chlorine water followed by 
washing with potassium hydroxide. The true incrusting 



1 E. Fremy and A. Terreil, Compt. rend., 66 (1868), 456-460; Bull. soc. 
chim., 9 (1868), 437-441. 



THE COMPOSITION OF WOOD 27 

substance was determined by difference. Oak wood had the 
following composition: 

Per Cent 

Ligneous cuticle 20 

Cellulose 40 

Incrusting matter 40 

The incrusting matter consisted of 10 per cent of substances 
soluble in water, 15 per cent soluble in alkali, and 15 per cent 
of acid bodies, soluble in alkali, formed by action of the chlorine. 
The ligneous cuticle (lignin) contained more carbon than cellulose. 

The substances present in vegetable tissues are subsequently 1 
stated to be: (1) cellulose bodies; (2) vasculose; (3) cutose; (4) 
pectose; (5) calcium pectate; (6) nitrogenous substances; and 
(7) mineral substances. The cellulose bodies, cellulose, 
paracellulose, and metacellulose, are distinguished from one 
another by their reaction to cuprammonium solution; cellulose 
dissolves immediately, paracellulose only after the action of 
acids, while metacellulose, which is found principally in fungi and 
lichens, is insoluble even after the action of acids. Vasculose 
cements the cells together and forms a large part of the vessels 
and tracheids; it contains more carbon and less hydrogen than 
cellulose, is rapidly attacked by oxidizing agents, and is insoluble 
in cuprammonium solution and 72 per cent sulphuric 
acid. Cutose, the fine membrane covering leaves and the green 
parts of plants, is soluble in alkali. Pectose after treatment 
with hydrochloric acid becomes soluble in water, from which it 
can be precipitated with alcohol. 

Fremy and Urbain 2 were of the opinion that a hard wood 
contained more lignin (vasculose) than a soft wood. Poplar 
contained 18 per cent of lignin and ironwood (bois de fer) 40 per 
cent. It is now known that there is no relation between hard- 
ness or specific gravity and lignin content. Vasculose was 
isolated by treating the wood with neutral solvents, dilute alkalis 
and acids, and exhausting with cuprammonium solution. The 
light-yellow residue had the appearance of the original tissue. 
Atmospheric oxygen in time changed it into a resinous mass soluble 

*E. Fremy, Compt. rend., 83 (1876), 1136-1141; 93 (1881), 926-931. 
2 E. Fremy and Urbain, Compt. rend., 94 (1882), 108. 



28 CHEMISTRY OF CELLULOSE AND WOOD 

in alkali. Through this modification of vasculose was explained 
the changes which wood undergoes in contact with the air. 
The compound celluloses have been classified as follows : 

1. Lignocelluloses contain lignin; wood, straw, jute, etc. 

2. Pectocelluloses and mucocelluloses contain pectic and 
mucilaginous or gummy substances; the former are represented 
by hemp, flax, and ramie, and the latter by algae and various 
fruits and tubers. 

3. Adipocelluloses and cutocelluloses contain waxy and fatty 
compounds. Cork, an adipocellulose, contains among other 
substances phellonic and suberic acids. The epidermis of the 
leaves and twigs of phanerogams contains cutin or cutose; on 
saponification with alkalis two principal fatty acids, stearocutic, 1 
C 2 8H 48 04, and oleocutic, Ci 5 H 2 o0 4 , are obtained. 

The above classification based on the researches of Fremy has 
the merit of convenience, but in a strict sense it is not accurate. 
Many fibrous materials would fall into two or more classes. 
Carried to a logical conclusion no simple cellulose exists in nature, 
since even raw cotton contains small amounts of cutin and pec- 
tin. The non-cellulose compounds are associated with cellulose 
in very variable amounts and there is little evidence of a chemical 
union with the cellulose. 

It has been a much disputed point whether in the wood the 
lignin is chemically combined with the cellulose, admixed, or 
adsorbed. Sachsse 2 compared wood with an alloy; it consisted 
of cellulose particles intimately incrusted by other substances 
originating from it or the cell contents. Von Baumhauer 3 work- 
ing with fruit pits supported the view of Payen that the non- 
cellulose constituents were mixed in different proportions with 
cellulose, though in peach stones the incrusting material was so 
intimately associated with the cellulose that it could not be 
removed with strong potassium hydroxide. The incrustation 
hypothesis was also favored by Schulze. 4 



1 E. Fremy and Urbain, CompL rend., 100 (1885), 19-24. 

2 R. Sachsse, "Die Chemie und Physiologie der Farbstoffe, Kohlen- 
hydrate und Proteinsubstanzen" (1877), p. 146. 

3 E. H. von Baumhauer, J. prakt. Chem., 32 (1844), 210. 

4 F. Schulze, Chem. Centr., 28 (1857), 321. 



THE COMPOSITION OF WOOD 29 

Kabsch 1 thought that lignin was a decomposition product of 
the cellulose and remained combined with it. Later Erdmann 2 
held Payen's incrusting material to be a definite compound 
chemically combined with the cellulose. The wood of the silver 
fir (Pinus abies L.) after treatment with hot dilute acetic acid, 
hot water, alcohol, and ether was called "glycolignose" and 
assigned the formula C3oH 46 2 i. The latter on hydrolysis gave 
sugar and 60 to 65 per cent of "lignose," Ci 8 H 2 60n. Lignose on 
heating with dilute nitric acid gave cellulose. Glycolignose was 
looked upon as a chemical entity, owing to the constancy of its 
composition and the inability of cuprammonium solution to 
remove more than traces of cellulose. Lignin was so difficult to 
remove from cellulose that Hoppe-Seyler 3 and others 4 thought 
that the two must be combined as an ester or ether. 

The chemical combination theory was early championed by 
Cross and Bevan. 5 Jute was assigned the formula (C 6 Hio0 5 )3.- 
C 6 H 6 3 and considered a definite compound since the fiber was 
attacked by hot alkalis and non-oxidizing mineral acids with loss 
of weight "but without producing any essential change in the 
insoluble residue"; it is not resolved by cuprammonium solution; 
and lignocellulose can be nitrated as such to give a definite 
tetranitrate. Muhlhauser 6 likewise obtained a definite jute 
nitrate behaving as a homogeneous substance. Treatment with 
alkalis left residues containing practically the same percentage of 
nitrogen as the original material. 

It is s.tated by Cross and Bevan 7 that, while the lignocelluloses 
are incompletely dissolved by such solvents as concentrated zinc 
chloride and cuprammonium solutions, and are incompletely 
precipitated from these solutions, there is no essential difference 
in reactions between the soluble and insoluble portions ; they have 
the empirical composition of the original fiber. In another work 

1 W. Kabsch, Jahrb. wiss. Botanik, 3 (1863), 369. 

2 J. Erdmann, Ann. Suppl, 5 (1867), 223-232. 

3 F. Hoppe-Seyler, Z. physiol. Chem., 13 (1889), 84. 

4 G. Lange, Z. physiol Chem., 14 (1890), 19; F. Czapek, Ibid., 27 (1899), 
165; V. Grafe, Monatsh., 25 (1904), 987. 

5 C. F. Cross and E. J. Bevan, J. Chem. Soc, 56 (1889), 199; cf. Ber„ 
24 (1891), 1772; 26 (1893), 2520. 

6 O. Muhlhauser, Dinglers polytech. J., 283 (1892), 88, 137. 

7 C. F. Cross and E. J. Bevan, "Cellulose" (1916), p. 114. 



30 CHEMISTRY OF CELLULOSE AND WOOD 

the question of a chemical union is left open. 1 The writer, 
employing cuprammonium solution 2 and the thiocarbonate 
reaction, 3 has found in the case of wood a decided difference in 
composition between the soluble and insoluble portions ; with the 
angiosperms especially the residues are much higher in lignin 
than the soluble portion. 

When wood is treated with 72 per cent sulphuric acid the 
insoluble residue shows under the microscope the identical struc- 
ture of the original material; for this reason Konig and Rump 4 
maintain that it is erroneous to assume a chemical union between 
the cellulose and associated compounds, and that the present 
classification of the compound celluloses can no longer be sup- 
ported. This argument against chemical union has little force 
when the behavior of some of the cellulose esters is considered; 
cellulose nitrate, e.g., in the form of threads and tissues can be 
denitrated to give a cellulose skeleton identical with the original 
in physical form. 

According to von Euler, 5 lignin is chemically combined with 
carbohydrates in wood, the union being that of a tannin glucoside. 
Karrer and Widmer 6 believe that cellulose and lignin are not 
chemically combined, since wood dissolves in acetylbromide to 
give a separable mixture of the decomposition products of cellulose 
and lignin. This argument is also without weight as the reaction 
is mainly hydrolytic; it will not take place in the absence of 
moisture or when the acetylbromide is free from hydrogen 
bromide. 7 

The formation of wood has been studied by Wislicenus and 
Kleinstlick 8 from a colloidal viewpoint. The cellulose hydrogels 
on coagulation form the cell tissues, which are thickened by 
adsorption from the sap of such substances as lignin, which is a 

1 C. F. Cross and E. J. Bevan, "Researches," III (1905-1910), p. 109. 

2 A. W. Schorger, J. Ind. Eng. Chem., 16 (1924), 143. 

3 Unpublished results. 

4 J. Konig and E. Rump, "Chemie und Struktur der Pflanzen-Zellmem- 
bran," Berlin (1914), p. 85. 

5 A. C. von Euler, Cellulosechemie, 3 (1922), 4. 

6 P. Karrer and F. Widmer, Helvetica Chim. Acta, 4 (1921), 700. 

7 L. Zechmeister, Ber., 56 (1923), 573. 

8 H. Wislicenus and M. Kleinstuck, Z. Chem. Ind. Kolloide, 6 (1910), 
17-23, 87-94. 



THE COMPOSITION OF WOOD 



31 



variable mixture. Esterification and other chemical processes 
may take place, but they play a minor part. 

Rontgen-ray spectrograms of jute and linden wood powder 
gave patterns differing from cotton cellulose only within the 
limit of experimental error; but whether lignin is adsorbed or 
chemically combined can only be determined by more extensive 
investigations. 1 

Constituents of Wood. — The major constituents of all woods 
are lignin, cellulose, and hemicelluloses; the latter comprise 
only those carbohydrates which are substantially insoluble in 
water. Ash, starch, and protein are always present in small 
amounts. The inclusion of pectin as a primary minor constit- 
uent is problematical. Wood contains variable amounts of 
secondary constituents, secreted from the living cells, which do 
not properly form a part of the wood complex. The major 
constituents are treated in separate chapters. 

Elementary Composition of Wood. — In 1844, Chevandier 2 
subjected wood from the trunk and branches of various species 
of trees to elementary analysis. The two forms of wood gave 
very similar results, the carbon content of the branches being 
somewhat the higher. The trunk wood had the following 
composition : 





Beech 


Oak 


Birch 


Aspen 


Willow 


c... 

H 


49.89 
6.07 

43.11 
0.93 
1.24 


50.64 
6.03 

42.05 
1.28 
2.05 


50.61 
6.23 

42.04 
1.12 

0.78 


50.31 
6.32 

42.39 
0.98 

1.86 


51.75 
6 19 




N 

Ash 


41.08 
0.98 
3.67 



The elementary percentages are based on the ash-free wood. 
The results for nitrogen and ash are too high. 

Analyses by Gottlieb 3 are given below. His values for nitrogen 
and ash are normal. 



1 R. O. Herzog and W. Jancke, Ber., 53 (1920), 2163. 

2 E. Chevandier, Ann. chim. phys., [3] 10 (1844), 143. 

3 E. Gottlieb, J. prakt. Chem., 136 (1883), 392. 



32 



CHEMISTRY OF CELLULOSE AND WOOD 



Species 



H 



OandH 



Ash 



Oak (Quercus pedunculata) . . . 

Ash (Fraxinus excelsior) 

Hornbeam (Carpinus betulus) 
Beech (Fagus silvatica): 

60 years old 

130 years old 

100 years old 

Birch (Betula alba) 

Pine {Pinus silvestris) 

Spruce (Pinus abies) 



50.16 
49.18 
48.99 

49.14 
49.03 

48.87 
48.88 
50.36 
50.31 



6.02 
6.27 
6.20 



43.45 
43.98 
43.31 



6.16 
6.06 
6.14 
6.06 
5.92 
6.20 



44.07 
44.36 
44.29 
44.67 
43.39 
43.08 



0.09 
0.11 
0.06 
0.10 
0.05 
0.04 



0.37 
0.57 
0.50 

0.54 
0.44 
0.64 
0.29 
0.28 
0.37 



Daube 1 attempted to distinguish between heartwood and 
sapwood by ultimate analysis. In pine only did the heartwood 





Age in years 




103 


104 


75 


125 


180 




Larch 


Pine 


Spruce 


Oak 


Beech 




Sap- 


Heart- 


Sap- 


Heart- 


Sap- 


Heart- 


Sap- 


Heart- 


Sap- 


Heart- 




wood 


wood 


wood 


wood 


wood 


wood 


wood 


wood 


wood 


wood 


c 


49.57 


49.86 


50.18 


54.38 


50.03 


49.55 


49.15 


50.28 


48.92 


49.06 


H 


5.85 


5.91 


6.08 


6.31 


6.05 


6.18 


5.84 


5.62 


5.86 


5.91 


N 


0.17 


0.12 


0.17 


0.56 


0.19 


0.18 


0.35 


0.28 


0.24 


0.22 





44.19 


43.99 


43.38 


39.00 


43.47 


43.89 


44.24 


43.66 


44.51 


44.41 


Ash 


0.22 


0.12 


0.19 


0.15 


0.26 


0.20 


0.42 


0.16 


0.47 


0.40 



contain more carbon than the sapwood, and this was attributed 
to the presence of resin. 

The leaching of chestnut wood to remove the tannins had 
little effect on the ultimate analysis, as shown by the following : 2 





C 


H 


O 


N 


s 


Ash 


Chestnut wood, normal 

Chestnut wood, leached 


50.28 
50.09 


5.58 
5.65 


43.21 
43.33 


0.10 
0.10 


0.03 
0.02 


0.80 
0.81 



It will be observed that wood contains approximately 50 per 
cent of carbon, 6 per cent of hydrogen, and 44 per cent of oxygen. 
Ultimate analysis shows no distinction between hardwoods and 
softwoods, or between sapwood and heartwood. 

1 W. Daube, Forstliche Blatter, 20 (1883), 177-192. 

2 H. S. Sherman and C. G. Amend, School Mines Quart, 33 (1911), 30. 



THE COMPOSITION OF WOOD 



33 



Analytical Composition of Wood. — Within recent years fairly 
complete analyses showing the composition of wood have been 
made (Tables 2 to 4). In general, hardwoods contain more 
pentosans, acetyl and methoxyl groups, and less lignin and pen- 
tosan-free cellulose than the softwoods (gymnosperms). Incense 
cedar is exceptional in having a methoxyl value comparable to 
that of the hardwoods, but its lignin content is very high (37.7 
per cent). Mesquite has a lignin content comparable with that 
of the softwoods, but the cellulose content is very low. Balsa, 
noted for its lightness, does not differ in composition from the 
average hardwood. 

The hardwoods contain more hemicellulose than the softwoods, 
the principal one being xylan. Mannan and galactan may be 
present in softwoods in considerable amounts, while only traces 
appear in hardwoods. Konig and Becker 1 (Tables 1 and 4) 
determined hemicellulose by heating the wood with 0.4 per cent 



Table 1. — Sugars in Hemicelluloses 



Sugar 



Fir, 


Pine, 


Birch, 


per cent 


per cent 


per cent 


26.0 


24.8 


61.1 


23.4 


21.4 


14.4 


3.4 


4.2 


3.5 


24.6 


43.4 


7.1 



Beech, 
per cent 



Pentose (xylose) 

Glucose 

Galactose 

Mannose 



73.9 

20.1 

0.1 

3.3 



sulphuric acid at arbitrary pressures: 2.5 atmospheres for fir, 
3.5 for pine, 2.25 for birch, and 1.0 for beech. Galactose was 
determined by oxidation to mucic acid, glucose by oxidation to 
saccharic acid, and mannose as the phenylhydrazone. Fructose 
was not found. 

Dore 2 has laudably attempted to account for all of the major 
constituents of wood. The contention is made that by the 
methods employed no constituents of importance are overlooked, 
and the overlapping of constituents is avoided. Undue weight 
may be placed on a summation approaching 100 per cent. 
Less information on the composition of the woods is available 

1 J. Konig and E. Becker, Z. angew. Chem., 32 (1919), 155. 

2 W. H. Dore, J. Ind. Eng. Chem., 12 (1920), 476, 984. 



34 



CHEMISTRY OF CELLULOSE AND WOOD 











NHO00 


CO 


CM*Ol>t* 


. IC 


rHi-HOO 


co 


>* 


■HH 




s 
















00 CO 00 IO 




OOOSC 


i CM 


i-H 00 CM 


o 


(N 


CM 














9SOini[9Q-X 


©00>OiO 


d 


CDCDOOC 


3 d 


COiOtN 




00 


oo 


HHHCOCO 


os 














CO CM CO CM 


co 




I CM 


CO rH t}h 


HH 


l-H 




HCMHCM 












•CMHHCM 


CD 


i-i OS CO if 


H CO 


(Nt^CM 


^ 


IO 


IO 


CDOsl>l> 


CM 












asopn a O-fi' 


■iQiOO 


>o 


3 O 


Tt<l>00 


CO 


os 


OS 


HCDHCM 


CO 














•lO!OC5 


d 


h©hC 


> i-l 


IOCMCO 




CM 


CM 


ocdosr^ 


CO 










o 




H 


i—i 


T-H i— I r-H r- 


H l-H 


T-H l-H 


l-H 






HrH CM 












OOt^kOO 


o 


G0CDN.O 


1 00 


NNO 


CM 


^H 


H 


00HHCS|> 


,_! 














i-Hi-HcOHH 


i-t 


00 OS CO if 


5 CO 




os 


00 


00 
















asoimpQ-w 
































CSCDI>iO 


CM 


CM CM OSW 


5 CM 


odi-HO 


CO 


00 


oo 


OHHHHOS 


^< 










a 




COCO>0>0 


CD 




3 CO 


•HHTt<lO 


Tt< 


1> 


l> 


l>COt^T)H 


CD 










MNOO 


00 


COH lOH 


1 00 


CO i-H CO 


OS 


§ 


ns 




iO 










h 


-u3diAq^ap\[ 


i—l OS OS OS 


Ci 


OOS SDK 


3 l> 


rHt^lH 


os 


o 


IOCS CO iO 


OS 


CMCOOOS 










CMHH.H 


H 


MHHr- 


H ,-H 


CM i-H CM 


■H 


csi 


CM 


HCMHrH 


H 


HHHO 


l-H 






ONOM 


CM 


OOJIOS 


3 O 


cooco 


oo 


o 


o 


I> 00 OS CD 


co 


00 OS OS© 


,-t 










IOCS CM CD 


00 




3 CO 


IOCS 00 


o 


•<J* 


TH 


THCOHOS 


co 


COHCOOS 


t^ 










OS 00 tHH^hh 


CD 


00 00 CD if 


3 l> 


OSOOS 


OS 


t^ 


t^ 


HHiOkOlO 


U3 


00t>t^.l^ 


t^ 






*OiOOSCM 


iO 


HNMt- 


• CM 


TjHCOt^ 


00 


00 IO 


F- 1 


CMiOiOCM 


HH 










uraSiq; 


t^00CNt> 


CD 


CMCM -*C 


q CO 


r-ll>i-H 


CD 


i-H CM 


CM 


CMHIO00 


HH 












CO»0 COt~ 


CO 


MHOC 


3 i-l 


001>t^ 


r^ 


tHtH 


HH 


t^HH|>CO 


co 














CM <N CM CM 


CM 


COCO coc 


3 CO 


co co co 


co 


coco 


co 


CM CM CM CM 


CM 










CM CM 00 CM 




io i* oor- 


• CO 


THOCO 


o 


t-^co 


>o 


COOSOH 




cocoHtN. 


00 








CM 1> 00 00 


TtH 


■tONr 


H 00 


OSi-Hl> 


CD 


COCN 


HH 


IOCMHHCO 


t^. 


co»ohhco 


HH 






osopixpQ 


cot^ooco 


r^ 


T-H T}H -HHlf 


3 CO 


OStHO 


,_, 


00 00 


00 


00 CM OS 00 


ns 




00 




CO 




iQiOiQkO 


iO 


iO "O »0 if 


3 >0 


COt«-*H 


H^ 


■*T« 


HH 


lOcDiOlO 


«o 


lOiOCC 


iO 


to 






OOTfHN 


CM 


l>t~t>b 


- CM 


OOCO 


»o 


!>■* 


iO 


HIOCOO 


CM 




O 




a 


UBSO^uadpfq^aj^ 


CO CM 00 t^ 


CD 


os io cs i- 


■1 TJH 


CM CO CM 


co 


t^l> 


t> 


H< CM CO OS 


CM 


hh CM 00 CD 


CO 


w 


a 

CO 

n 




i-H i— 1 i— 1 T-H 


i— • 


COCO CMC< 


3 CO 


l-H T-tl-H 


l-H 


CM CM 


CM 


CO CO CO CM 


co_ 


CO CO CO CO 
CO CM l> CO 


co" 


Q 




CO CD 00 CM 


lO 


cocMCMt* 


■ t^ 


1>OSO 


>o 


cot> 


o 


IOCS 00 CD 


CD 


O 

o 

> 




OCSOOlO 


CO 


O0iOl>C 


3 00 


OOSJ> 


CD 


OS CO 


00 


l>rH HHtHH 


OS 


eoioiOHH 


HH 


urcso^ua^ 


00O3iO>O 


t> 


00 00 COb 


• I> 


IHOSO 


d 


l>l^ 


t> 


coi>coj> 


CD 


l>t^l>t^ 


t^ 






U5NHN 


OS 


NOiOC 


3 IO 


■THOSOS 


">* 


r^co 


H 


00OSCOH 


CD 


OOCDCO 


IO 


V, 




to 

O 


pfxoqiap\[ 


wcooo 


TjH 


hh COOC 


S <N 


COCMO 


CN 


CM i-H 


CM 


COIOOOHH 


IO 


OOS CMO 


o 


«*1 




■HH ^ TjH H/ 1 


rJH 


IQIOOT 


H IO 


co co co 


CD 


IO IO 


iO 


HHHHHHTfH 


•^i 


lOrflOiO 


IO 


n 




CM HH CO 00 


OS 


CO CM lOT 


' OS 


cothco 


,-H 


coco 


on 


HHHl>OS 


co 


OCOCMOS 


CO 


fa 


v - " 


pioB oi^soy 


OjNOh 


o 


CO CO IO f 


3 IO 


aoh 


OS 


OrH 


o 


CM OS COO 


O l>OSCOt^ 


t^ 


t 




dni-HH 


1—1 


1— 1 1— 1 1— 1 1- 


■1 i—l 


OHO 


d 


l-H l-H 


rH 


l-HOHr-< 


l-H 


ooc 


c 


d 






00 00 CO l> 


o 


OCSCSt> 




t^THOS 


OS 


CDtH 


o 


NONH 


00 


CMI>CT 


CO 


T3 




HO^N 


COhM 


CO 


0Cri< COCC 


3 Tft 


CO 00 00 


CD 


OOS 


o 


OS t^OS »0 


I> 


coooooo 


CO 


<1 


d 




^uao aad x 


CM CO CO OS 


o 


CO HH CMC 


) CO 


l>COl-H 


1^ 


OOS 


o 


CCCMIOCO 


HH 


iH'rHOOH' 


CM 


fa 

O 




<NH«H 


CM 


r-t H i— It- 


H l-H 


1-HrHCM 


1—1 


CM i-H 


CM 


l-H T-H l-H l-H 




CMCMHCM 


OJ 


r> 


P 




CO CM IO 00 


CM 


t>HH OOr- 


^ iO 


COOSO 


r-H 


OtM 


h- 


OCMOOCO 
OHHOSCD 


CD 


oo>oco 


CM 


o 






COIOHH-hh 


iO 


COCOOi- 


I iO 


co co os 


CO 


OH 


o 


CM 


l>t>CDCM 


co 


co 
fa 


>i 


J3q^3 


OS CD OS CO 


00 


CMCO CMC 


i CM 


co^^ 


TH 


HH 


H 


HHHH CO HH 


^ 


CD CD CMO 


CO 




l>OCOOS 


>o 


OS CO CDC 


l-H 


i>CMTt< 


GO 


l>TfH 


CD 


lOl^iOOO 


OS 


OCOOOl> 


iO 


0!i 


O 
bC 
<& 
-p 
d 
a> 
o 

S- 

a; 
d 


x> 




CD HH CO 00 


o 


OOHHOOCN 


l-H 


00 CO CD 


co 


I>OS 


00 


COiOCMl> 


HH 


CMOI>iO 




Hi 

<1 


S3 

'o 


ja^BM. ^OJJ 


IO CO CO HH 


>o 


CMCOCMC 


CO 


T)<COt^ 


iO 


OS OS 








00 CD CD t~ 


l> 


£ 




O0l><Mt> 


OS 


CDCOt^T 


t^ 


OS CO i-H 


th 


HO 


CD 


OCOCMO 


CD 


iooo»o 


o 






IOOSCOH 


o 


COOHHt> 


T^ 


©lOCO 


CD 


CO-* 


co 


COt>OS-<H 


i-H 


l>COHHO 


CM 


<! 
| 




aa^BAv pp3 


»0 CM hh CO 


■* 


HCOCMCs 


<N 


COCMiO 


co 


t^l> 


t> 


CMHCOHH 


CO IviOiOcO 


CD 






COCM»OHH 


CO 


CO"OCM 


• CO 


00 1^00 


-tl 


CMO 




HOOOOS 


O 


OHHiOOO 


r^ 


CM 


qsy 


hh-hhiOHH 


■* 


CO CO CO 


• T« 


CO CM CO 


co 


CM CM 


CM 


CMCMHrH 


CM 


HH CO CO CO 


CO 


fa 


OOOO 


o 


OOO 


o 


OOO 


o 


do 








OOOO 


o 




CM 00 OS OS 


CM 


-tf ostei- 


1 OS 


CM CO CM 


CM 


tH^h 


oo 


OOOOOiO 


O 






3 




am^siop\[ 


COCMOOH 


Th 


te i-H aoo 


00 


eocc* 


1-1 


coi> 


CD 


HCOOH 


o 






CO 


lOOOCOiO 


CO 


COlO rUtf 


) T* 


IO tH iO 


>o 


OOS 


OS 


coi>t^t^ 


t^ 










Pi 









a 




1=1 




(3 




a 






HCMCOHH 




i-l CM COT 




lOCOt^ 




00 00 


03, 


HCMCOHH 


cs- 


HCMCOO 


§ 










coco co cc 


a> 


CD COCO 




CD CD 


0J 






HHHIM 




aidui^g 




s 


T ~" H '~"~ 


§ 


rirtri 


§ 


l-i,-\ 


a 




% 




s 




' 
























"e 








^ 




OS 




'e 
















© 




J 




s 




60 




o 
u 
















■8 
e 

o 
0, 




8 




s 




s 

s 




s 




99 












•5 








'a 
s 




s 








69 










£ 




SO 




^ 




'O 




Oh 




a 








os 






e 
















69 








02 


G 
'ft 




s 

e 






o 
u 




1=1 
ft 




3 

5 










£ 








_, 




s 








„ 






















rG 



















>> 

a 

1 
1 




eS 












XI 




a 












T3 
a 

Tj'S 




o 

(V 

a 






) 


u 

V 




ft 
1 
g 












* 






a 






* 




a 







THE COMPOSITION OF WOOD 



35 























■^•"^COl-H 


oo 


1>C0^C0 


r^ 


on 


00 


0)H 


iO 


THlOrH 


© 














©rHCM CM 


o 


CMt^lOrH 


rH 


o 


o 


OCO 


co 


©CM© 


1-1 
















oo 


NOHO 


rH 


H^ 


•* 


CM 00 


o 


rHrHrH 


rH 














CM-frHCM 


CM 


CM CM CM CM 


CI 


CM 


CM 


CMrH 


CI 


co co oo 


00 














— dOCMt~ 


(M 


CO CO CO HH 


Ui 


r^ 


t^ 


■*H 


CM 


rHOO 


© 














00OOCM 


o 


NiOOO 


co 


CM 


CM 


CD© 


(/J 


rHOO 


l> 














00 CM CM 00 


o 


rH CM CM CO 


CM 


© 


O 


CM CO 


CM 


CMOO 


o 














rH NH 


rH 






























OHiOCN 


1^ 


OHOM 


on 


HH 


-* 


t^oo 


CM 


10*0 


CO 














lOOSHCO 


t> 


Ol>HH00 


HH 


CD 


CO 


CM CO 


OO 


00l>O 


oo 
















CO 




CO 


IO 


iO 


lOI> 


© 


l>O00 


or, 














IO iO iO iO 


>o 


«>1>1>I> 


t^ 


l> 


t^ 


t^t^ 


t- 


CO©© 


© 


to 10 


o 


©cMhh© 


TO 


0000 CO 


CM 






t^rH -CO 


rH 


IO 


IO 


t 1 © 


rH 


CM-* H-cO 


© 


<N «H 


CM 


-HCMCM© 


rH 


OOCO CO 


r^ 






©CM -TO 


00 


CO 


CO 


COrH 


TtH 


©■-tfCMCM 


HH 


«c 


r-l 


rH 


HHHO 


rH 


oo o 


o 






rHrH -O 


© 


rH 


rH 


rHrH 


rH 


CO CM CMrH 


CM 


CM 


HH 


CMrHt^lO 


Tt* 


COO CO 


00 


CMOOCM 


(M 


OTOt^HH 


iO 


TO 


TO 


COlO 


TO 


IOCMO© 


© 


lO i-i 


CO 


tHHhcoiO 


OJ 


CMCM 00 


CO 


CM-* CO CO 


00 


OTO»OHH 


1> 


TO 


TO 


COHH 


CO 


CO©rHt^. 


TO 


»o lO 


iO ©00 00© 


00 


OO O 


o 


COiO©(N 


CM 


l>t^l>l> 


r^ 


TO 


O 


rHCM 


rH 


OrHOrH 


O 










!-H 




CM CM CM CM 


CM 


rH rH rH rH 


rH 


r-l 


y-< 


CMCM 


CJ 


CM CM CM CM 
















ONOCO 


>o 


1> 00 00 CM 


r^ 


CMt^ 


O 


oo^ 


rH 


■*HHt>Tf 


^ 




















CMOCM00 


(JO 


CM rH CM CM 


TfH 


IO Tt 


IO 


00© 


<* 


©CMOt^ 


O 




















COCOiOHH 


-H 


OOrH© 


© 


CO CD 


CO 


coco 


co 


TfH lOlO © 


iO 




















CM CM CM CM 


Oi 


oo co co co 


00 


CMC 


CM 


CMCM 


CM 


CM CM CM CM 


CM 


t>O00CM 


r~- 


l/J rH i— 1 rH 


o 


rHoscoo 


IO 


OOMO 
■HiOMOl 


CO 


O001>l> 


oo 


H4-H 


>o 


©iO 


CM 


t^coiooo 


CM 


O1OO00 


tJH 


CMt-OCO 


00 


COCO -*o 


oo 


© 


t-CMOOO 


HH 


©C 


^ 


©00 


CM 


©iOTt<00 


© 


HNMM 


rH 


00 00©CO 


l>« 


CMCOOrH 


rH 


O©t>00 


oo 


TfHlOlOlO 


IO 


f f 


■* 


lOtD 


CO 


000©iO 


r>- 




IO CO CD 


CO 


IO io CDiO 


IO 


CO COCO CD 


CO 


1O1O1O1O 


>o 


TH rf< HH HH 


Tfl 


lO'C 


iO 


iO"0 


IO 


iO IO »o iO 


IO 


■^00 00 


rH 


1>C0t*O 


rH 


OOCMIO^ 


iO 






OOCO-^ 


© 


00 cc 


CD 


CMt^ 


© 


I>*Tt<00 


oo 


CM©COCO 


TH 


HH 00 rHOO 


oo 


COOICO 


io 






lOCOrHCO 


t^ 


oox 


00 


OO 


00 


OrHt>.HH 


co 


"c- 


■■* Tf* <# 


■* 


CM CM CO CM 


CM 


CO CO CO CO 


00 






OOHO 


© 


©c 


© 


o© 


o 


rHCM CM CM 


CM 


oooo 


CM 


io*cqoo 


o 


00— IHHCM 


o 


CMOCOCM 


TO 


HHCJIOIO 


CD 


rHCT 


iO 


00 CO 


CM 


rHCDO© 


TO 


OWON 


O 


HONN 


00 


t^COO-* 


oo 


OOH« 


LO 


OTO00TO 


TO 


lOt^ 


co 


IOO 


00 


HHCOOCO 


© 


CO©©IO 


co 


HHOO 


o 


oooo 


o 


oooo 

CMCMrHrH 


TO 


"* CO CO CO 


CO 


t^r- 


t«. 


00© 


00 


rHOJ>© 
CM CM rHCM 


© 
CM 


rHt^OOCM 


IO 


00rH00lO 


co 


rHCOTOCM 


o 


TfffllOOl 


-H 


00TOTOCO 


iO 


HIT 


00 


VHIO 


co 


rHl>I>CO 


oo 


00 rHOO© 


© 


©TO©© 


o 


CO CM CM CO 


oo 


COrHCOO 


l> 


CD CO CM CO 


'O 


l>CC 


CO 


coco 


© 


rHOO 00 IO 


t^ 


-hiOt*,* 


tH 


lOHHlOlO 


>o 


iO lO >OiO 


iO 


lO©iO© 


iO 


lOiOiOlO 


>o 


LO>r 


IO 


lOiO 


>o 


!>©©© 


CO 


C0rHC0l> 


<* 


HHCDIO 


rH 


00t>©00 


OS 


©ooco 


00 


COOOOIr^ 


co 


iO»r 


© 


©CM 


rH 


rHl>rHCM 


IO 


»oho 


o 


©ot^ io 


o 


lOiO^N 


io 


l>l>Tj<rH 


CM 


lOt^iOCO 


© 


»>oc 


00 


©■*! 


IO 


COTOiO© 


oo 


c 

C\ 




rH 


oooo 


o 


H rHrHrH 


rH 


ICIOHHIO 


iO 


rHrHCNCN 


CM 


lOlT 


IO 


CMCM 


CM 


CM rHrHrH 


rH 


©CMCM 


rH 


t>00HHl-| 


Tt< 


CO 00 iO CO 


h- 


OOOrHCO 


CO 


CM00OTO 


CM 


r^cc 


o 


IO-* 


"H 


b-CMO© 


r>- 


Xt»HN 


T-* 


OO^H 


<-< 


C0iOl>CO 


IO 


lOCOOO 


TO 


!>©COCO 


10 


coco 


CO 


©■>* 


© 


lO-^oi-* 


IO 


iO©iO© 


CD 


(NHOK5 


CM 


rHrH CM© 


rH 


CM »0 CO CM 


oo 


OOOOlO 


OO 


oc 


© 


00TO 


TO 


©OOt^rH 


GO 


rH I— 1 1— 1 i— I 


rH 


CMCM — (CM 


CM 


tH rH rH rH 


rH 


CM CM CM CM 


CM 


CM CO CM CM 


CM 


CJC 


CM 


i-i i-f 


t-< 


rHrH rHCM 


r-( 


tHOiH(N 


(N 


CMTt<cOCO 


rH 


>OONM 


CD 


H<C0O00 

t>t>ooo 


© 


O -cot- 


© 


r^<^ 


CO 


lOCM 


co 


rt<CMOO 


CO 


OlOHO 


O 


t~l>©00 


oo 


©Olfflffl 


00 


00 


cm -coco 


co 


CM — 


CM 


coco 


© 


IO<0©>0 


iO 


OHHH 


rH 


oooo 


o 


rtOOH 


rH 


o©o© 


o 


CM CMCM 


CM 


rHr- 


i— 1 


©O 


© 


©o©© 


© 


cm t> coco 


o 


NO00O 


O 


00 00 CM 00 


~V 


lOiOCMOO 


o 


t^t^cOt^ 


TO 


H. H- 


o 


rH 00 


r^ 


rHCD©l> 


00 


©OOOO 


io 


ic^on 


IO 


oocM>oao 


** 


CMrHTO© 


CO 


CMI>»01> 


© 


oot^ 


1> 


rtH>. 


iO 


T^OCMCM 


o 


©©© O 


CO 




























© 


o©hhio 


HH 


loococo 

HOH00 


rH 


00 CM iO CM 


C) 


hh CM CM CM 


o 


•OO00rt< 


CI 


ioot 


t^ 




oo 






l>rHTO<M 




CO 


CMTO^OO 




rHC0CMt> 




lOiOcOO 


CD 


OC'CC 


I> 


t^00 


l> 


©O00l> 


CD 


CO CO CM-* 


CO 


OrHOOCM 


o 


rH©rH© 


M 


"HH * ■«* CO 


-* 


CM 00 CM rH 


CI 


rHr- 


1— 1 


TjHHH 


Tt< 


CMHHiOlO 


tT 


Ot^iOOO 


00 


rHCM CM© 


CO 


cotoocm 


rH 


OOrHIMrH 


co 


l>00TO© 


-V 


lOC 1 


CM 




TO 












C4 


CO CM CO CO 


OO 


t»00 00O 


oo 


IO iO hh io 


«o 


HC 




COt> 


© 


CM CM CM CM 


CM 


oooo 


o 




























O 














O IO CO CO 


CD 


OOOiO© 


TO 


o-t 


l> 


©TO 


TO 


TOiOt^O 
















rHOJCMOO 


CD 


ICrHrHCM 


H- 


>o-t 


tH 


©CO 


-h 


OiOOOO 


iO 














Ttioocooo 


CO 


TticOcOlO 


iO 


icco 


CD 


ON 


oo 


©©©IO 


CO 


rHCM CO IO 


c3 


—KNCO"* 




rHCM CO"* 


eS 


HOimn 


a 


rHCNOO-rt* 


a 


iO»C 


a 


t-00 







c 




0) 




<D 




QJ 


lomiouj 


a> 


t>l>t-t> 


n> 


l>-N 


CD 


t^t^ 










2 




s 




£ 


. 


X 


HHHri 


2 


rH r- 


£ 




% 




£ 
































'a 








s 




!■£ 








e 














§ 




<o 






8 

1 




CO 

s 

ft) 




^ 




£ 










.e 




3 

►o 




e 




^ 




2 


















ia 




1 




1 

3 








s 
s 




."a 

00 

o 




5 

e 

c 
c 


> 


H 

c3 




a, 








s 




1i 

2 




M 
o 




I 








j2 

rfl 
OQ 




CO 






00 

eS 

o 

Q 




s 

-t- 



1! 


1 
> 






a 

IE 


H 




i 
s 




5 
I 
% 


t 




T 




o 
M 

o 

w 




a 

c 





36 



CHEMISTRY OF CELLULOSE AND WOOD 



o 

s 


asojniiaQ^ 


















asojnjpo-tf 










asojnjpQ-*) 












HTfOTfO 


TfH rH CM rf 


CO 


"># LOOt> 








IO HC0O 




ooot^ 


o 




-u9djXq;aj\[ 


■-H-H— I^H!M 


rH rtrtrt 


_; 


^^^6 


o 




M^HJIO 


X OCDlO 


o 


X CM COO 


X 






i»ocaox 


<N TfOO 




©XXCM 


■* 




u^soiuaj 


■sfcoco^Tf* 


tP XC3CO 


X 


i— 1 LO LO LO 


"* 






NNJ1MN 


CM CM CM CM 


cm 


cm cm cm cm 


CM 


uiuSrj 














NhC^X 


*# o x o; uo 


rH 


XNoa) 


X 






O^CriO 


CM -*OXX 


co 


1>cDCMtP 


l> 




asopiipo 


CM CM rf co co 


ih d^'^^ 


,_! 


OhOO 


o 






co co io co to 


CO CO CO CO CO 


CD 










CM LO X CO CO 


CO X CM >-0 rH 


03 


C — LO LO 


o 


UBSOiuadpSq^aj^ 


NXSrHCl 


i> i-i.-hcmcm 


CO 


ChOX 


co 






CO CO CO Tf* CO 


co co co oi cm 


07 


CM CM CM CM 


Ol 






CM ■* l> -~ OS 


CO CO O O CO 


CO 


OOiHCN 


,_i 




UBS01U3J 


XiOCOrtN 


O CM'C-OX 


ec 


HONCC 


t>. 




OCOSO 


O ■* LOCO 1-0 


— 


l-fi-iCMrH 


^H 






— 1 — iCM^hCM 


rH CM 0) N -1 


01 


tN 01 Ol 01 


N 






CO LO ^H rH CM 


O CM CO 03 -N 


h- 


cmcolox 


LO 




[^XOt^3J\[ 


NOHCN 


O HOHO 


o 


CM CM CM CM 


Ol 






CO CO CO IO IO 


CO CO CO CO L0 


CD 


|>1>I>I> 


t> 






X-*CC — IX 


OS 03 OS iH CN 


— 


^ i_~ 3 — 


CO 




pioB ouaoy 


t^HTfTfiH 


I> OCOXO 


co 


CM CM CO t> 


">* 






lo co lo io eo 


lo co — co io 


"tf 


T^^^T}' 


■sjl 






co — I co w x 


CO NOH lo 


LO 


XOr-LO 


■* 




HO«N 


■<*oc5Trco 


t^ O CM LO CD 


X 


OS CM CCO 


CO 




^U3D aad x 


co ^h o — i lo 


co OOOO 


OS 


CCNXX 


t> 


£ 




N CN 01 CN N 


OI M 0) — 1 rH 


— 


i— i i— 1 1— 1 1— I 


■H 




O ■* 03 03 X 


co lo t^ — co 


o 


CCMOO 


LO 


>> 


•iaq}3 


IO iH IO 00 CO 


O LO CO LO CO 




CM CM CO CM 


CM 


-IH^ON 


-h OOOO 


o 


OOOO 


O 




■* ■■# co ex i> 


b- — t^ CO LO 


r^ 


I> CM X LO 


CO 






XC! OCMCO 


O N 00 CO iH 




CM CM I> i-i 


CO 


X 


jajBAv ;og 


COCNUOCOTf 


tjh ->*COTtCO 


CO 


^ Tfl ^ Tf 


-* 




tjh co ■* co lo 


IM X X lo cc 


r^ 


oco — co 


LO 






O co tH cm lo 


HOCttjHO 




coi>c-co 


CO 




laiB.w pioq 


CM^CO»-lCM 


CM CM CM CO CM 


(N 


CM CM CM CM 


CM 






O TT" CO — L~ 


CO XN-h 


0) 


CO — OX 






qsy 


X I> 03 OS X 


X LOLOLOCO 




TTLOTf CO 


"* 




ooooo 


O CCOO 


O 


CCOO 


o 


am;sioi\[ 
















c 


c 




— 






iHNM^iO 


C3 t-ICMCO^ 


cit 


'-HCMCO-* 






9[dUIBg 




o 






4> 
















1 










1 

3 


1 




e 






03 


s 










O 


a 




C3 






© 


s 










a 






so 






02 


O 


CJ 

tH 

3 




s 

U 








O 

1 


jo 










PC 




> 






00 







o co 



-•co 
r bio 

g 5^lo 

^3 ■ l> 

• eo 

■«• K - 

°£ 

"2 5« 
ca Q 

a c a 
a o o 

r- < O 

1^ 



THE COMPOSITION OF WOOD 37 

Table 3. — Analyses of European Woods 1 



Spruce 

(Picea 

excelsa) , 

per cent 



Pine 

(Pinus 

silves- 

tris), 

per cent 



Beech 
(Fagus 
silva- 
tica), 
per cent 



Birch 
(Betula 
verru- 
cosa), 
per cent 



Poplar 
(Populus 
tremula), 
per cent 



Ash 

Fat (a) Ether extract 

Wax (b) Alcohol extract. . . 

(c) Sum of (a) and (6) 

Resin (d) Alcohol-benzene 
extract 

Methyl 

Pectin (von Fellenberg) . . . 

Acetic acid (Schorger) 

Protein (N X 6.25) 

Furfural 

Pentosan 

Methylpentosan 

Cellulose containing pen- 
tosan 

Pentosan in cellulose 

Cellulose corrected for 
pentosan 

Lignin (hydrochloric acid 
method) 



0.77 
0.78 
1.52 
2.30 

2.34 
2.36 
1.22 
1.44 
0.69 
7.49 
11.30 
3.00 

63.95 
9.55 

57.84 

28.29 



0.39 
1.92 
1.53 
3.45 

3.32 
2.20 
1.11 
1.40 
0.80 
7.04 
11.02 
2.23 

60.54 
11.27 

54.25 

26.35 



1.17 
0.31 
1.47 

1.78 

1.20 
2.96 
1.75 
2.34 
1.05 
14.90 
24.86 
1.02 

67.09 
20.35 

53.46 

22.46 



0.39 
0.71 
1.09 
1.80 

1.68 
2.77 
1.61 
4.65 
0.74 
16.08 
27.07 
0.84 

64.16 
29.40 

45.30 

19.56 



0.32 
1.08 
2.08 
3.16 

2.87 
2.57 
1.82 
4.17 
0.63 
12.64 
23.75 
0.72 

62.89 
24.94 

47.11 

18.24 



C. G. Schwalbe and E. Becker Z. anaew. Chem. 32 C1919). 229-231. 

Table 4. — Analyses of European Woods 1 
(Results in per cent of dry weight) 





Pro- 
tein 
(N X 
6.25) 


Resin 
and 
wax 


Ash 


Total 
pento- 
san 


Hemicelluloses 


Lig- 
nin 


Cellulose 2 


Species 


Hexo- 

sans 


Pento- 
sans 


Crude 


Pure 


1. Fir 


1.21 
1.21 
1.27 
1.29 
2.29 
1.39 
1.14 
1.58 
1.30 
1.17 
1.89 


2.83 
1.71 
3.17 
2.47 
1.88 
2.66 
2.32 
0.70 
2.24 
2.04 
2.83 


1.10 
0.42 
0.53 
0.68 
0.46 
0.84 
1.21 
0.96 
0.83 
0.83 
0.49 


11.48 
11.63 
10.80 
25.86 
24.01 
22.71 
21.88 
24.30 
23.68 
23.31 
22.94 


13.58 
13.00 
12.78 
4.61 
5.00 
2.60 
3.43 
4.36 
5.70 
5.05 
3.65 


8.67 
9.74 
8.70 
23.20 
21.48 
15.36 
15.10 
17.79 
19.29 
16.75 
15.90 


29.17 

27.98 
29.52 
23.27 
26.38 
22.45 
20.75 
22.69 
26.01 
24.70 
24.57 


43.44 
45.95 
44.01 
44.52 
42.50 
54.71 
56.06 
51.93 
44.64 
49.46 
50.69 


40.62 


2. Fir 


44 06 


3. Pine 

4. Birch 

5. Birch 

6. Poplar 

7. Poplar 

8. Beech 

9. Ash 

10. Willow 

11. Alder 


41.93 
41.85 
39.97 
47.36 
49.27 
45.41 
40.24 
42.91 
43.64 



1 J. Konig and E. Becker, Z. angew. Chem., 32 (1919), 155-160. 

2 Crude cellulose contains pentosans; pure is free from them. 



38 



CHEMISTRY OF CELLULOSE AND WOOD 



in Table 5 than by the usual method of analysis. While being 
minor constituents, a summative analysis should include ash 
and protein. Information on the total pentosan content is 
lacking. There is also overlapping of mannan, for example, 
some of which remains in the Cross and Bevan cellulose. The 
extract with 5 per cent sodium hydroxide is meaningless as to 

Table 5. — Composition of Wood 
(Per cent of dry weight) 



Red- 
wood 



Western 

yellow 

pine 



Sugar 
pine 



Live 
oak 



Benzene extract 

Alcohol extract 

Soluble in cold 5 per cent NaOH 

Soluble in cold water 

Cellulose 

Lignin 

Pentosans not otherwise accounted 

for 

Mannan 

Galactan 



0.34 
4.39 



54.89 
34.50 

3.67 
3.21 
0.50 

101.50 



2.22 


2.84 


1.49 


1.90 


57.72 


59.18 


29.47 


29.50 


3.49 


1.86 


6.37 


6.63 


0.78 


0.50 


101.54 


102.41 



0.52 

4.52 

19.53 

3.82 
47.47 
21.14 

1.97 
0.00 
1.56 

100.53 



composition. The extract is a mixture of a portion of the pen- 
tosans and other carbohydrates, acetic acid produced by hydroly- 
sis of the acetyl groups of the lignin, protein, tannins, and other 
substances which give the lignin reaction. If a summation is 
desired it will be found by reference to Table 2 that addition of 
ash, hot-water-soluble, ether extract, acetic acid, pentosan, 
lignin, and cellulose (free from pentosan) gives a figure close to 
100. 

According to Klason, 1 spruce (Picea excelsa) has the following 
composition : 

Per Cent 

Cellulose (free from pentosans) 53 

Hemicelluloses 15 

Lignin 30 

Fat, resin, protein, etc 2 

1 P. Klason, Svensk Pappers-Tid., 24 (1921), 7. 



THE COMPOSITION OF WOOD 39 

Von Euler 1 gives the composition of pure coniferous wood 
substance as follows: 

Per Cent 

Cellulose 54. 10 

Hemicelluloses : 

Hexosans 3 . 06 

Pentosans 12 . 25 

Lignin, including acetyl groups 30 . 60 

The compostion of African wattle wood 2 (Peltophorum africa- 
num) is not strikingly different from that of American 
hardwoods. 

Effect of Storage on the Composition of Wood. — Time is of 
little influence on wood kept in a dry state. Schwalbe and 
Becker 3 were unable to detect any appreciable difference by 
chemical analysis between the wood (Picea excelsaf) of an Amati 
violin 260 to 280 years old and the wood used for violins in 
Germany at the present time. 

Sawdust from Scotch pine stored for 2 years showed no 
appreciable change in ultimate composition. The wood had 
apparently undergone oxidation since the air from a pile of 
sawdust stored for 1 year contained 6.3 per cent by volume of 
carbon dioxide. 4 It should be remembered that the gas in the 
wood of a living tree is high in carbon dioxide. 5 Bergstrom 5 
found 6.2 per cent of carbon dioxide and 15.4 per cent of oxygen 
in the gas taken from the sapwood of a growing Scotch pine 
and 5.0 per cent of carbon dioxide and 16.7 per cent of oxygen 
from that of a spruce. 

Relation between Density of Wood and Its Cellulose Content. 
Griffin 6 concluded that there was no relation between the number 
of annual rings per inch and specific gravity, or between the 
specific gravity and yield of cellulose. It is a general rule, how- 
ever, that the greater the number of rings per inch the greater will 
be the specific gravity. On a volume basis the wood with the 

1 A. C. von Euler, Cellulosechemie, 4 (1923), 1. 

2 E. F. English, J. Soc. Chem. Ind., 43 (1924), 742B. 

3 C. G. Schwalbe and E. Becker, Z. angew. Chem., 33 (1920), 272. 

4 H. Bergstrom, Papier-Fabr., 22 (1924), 37. 

5 J. Bohm, Landw. Vers.-Sta., 21 (1878), 372-388; H. Devaux, Compt. 
rend., 134 (1902), 1366-1371. 

6 M. L. Griffin, J. Ind. Eng. Chem., 6 (1914), 560-561. 



40 



CHEMISTRY OF CELLULOSE AND WOOD 



greatest specific gravity will contain the most cellulose by weight. 
Wahlberg 1 found that the yield of pulp per solid cubic meter was 
125 and 263 kilograms respectively for two spruce trees grown 
under apparently identical conditions, and 220 kilograms for a 
pine tree grown on poor soil. The percentage of cellulose based 
on the weight of dry, resin-free wood was very uniform. 

The relation between the number of rings, specific gravity, 
and yield of cellulose has been investigated by Wahlberg. 2 
The line of demarcation between the heartwood and sapwood of 
spruce is best developed by the use of a 1 per cent solution of 
osmic acid. The line did not extend equally in all radial direc- 
tions, but the differences did not include more than one or two 
annual rings; on the other hand, it appeared from the twenty- 
third to the fifty-first ring in various trees. The following table 



Table 6. — Relation between 


Width of Rings 


and Density 






Average width 




No. 


Annual rings 


of annual rings, 
millimeters 


Specific gravity 


72 


59-54 


1.6 


0.434 


73 


53-49 


1.8 


0.408 


74 


48-44 


3.4 


0.395 


75 


43-40 


2.9 


0.357 


76 


40-38 


6.2 


0.307 


77 


38-35 


4.8 


0.333 


78 


34-31 


4.2 


0.345 


79 


30-27 


5.1 


0.341 


80 


26-23 


6.8 


0.353 


81 


23-10 


4.1 


0.341 



shows clearly that the wood with the narrowest rings has the 
highest specific gravity. The apparent discrepancy between 
width of ring and specific gravity in Nos. 76 and 80 is probably due 
to the presence in the latter of a greater proportion of dense fall 
wood. A section of wood after drying in vacuo had a specific 
gravity of 0.345, while the average values for spring and fall 
wood were 0.307 and 0.601 respectively. From a comparison of 

1 H. E. Wahlberg, Papier-Fabr., 20 (1922), 1216-1218. 

2 H. E. Wahlberg, Zellstoff u. Papier, 2 (1922), 129-134, 155-164, 202- 
212. 



THE COMPOSITION OF WOOD 



41 



the density of certain rings and the weather for the paticular 
years which they represented, it is concluded that cold and wet 
weather leads to the formation of wood with a low specific gravity. 
This opinion cannot be generally accepted, as optimum growing 
conditions usually result in wide annual rings, hence a low specific 
gravity. No pronounced difference in ash content between 
spring and fall wood could be observed. 

The cellulose in spring and fall wood was determined by a 
preliminary digestion with bisulphite solution, followed by 
treatment with bromine by Miiller's method. It will be observed 
(Table 7) that after removal of the resin there is little difference 
between the cellulose contents of spring and fall wood in the 
trunk. The variations are within the experimental error of the 
methods employed. 

Table 7. — Cellulose in Spring and Fall Wood 



Annual rings 



Cellulose in wood 



Unextracted, 
per cent 




Extracted with 
alcohol-ben- 
zene, per cent 



Spring wood, 60-58 
Spring wood, 47. . . 
Spring wood, 55-54 
Fall wood, 55-52... 
Fall wood, 48-46... 
Fall wood, 34-33... 

Limb 

Limb 



47.6 
49.0 
47.9 
49.5 
47.2 
48.1 
42.1 
42.9 



Heartwood and Sapwood. — The effect of age on the chemical 
composition of alder wood (Alnus glutinosa) has been studied 
by Schwalbe and Becker 1 (Table 8). The differences between 
heartwood and sapwood are slight. The youngest wood 
contained the most protein, pectin, acetic acid, and pentosan, 
and the least cellulose and lignin. 

Cieslar 2 determined the effect of various growth conditions on 
the lignin content of several conifers. He used the method of 

1 C. G. Schwalbe and E. Becker, Z. angew. Chem., 33 (1920), 14-16. 

2 A. Cieslar, Mitt, forstl, Versuchsw. Oesterr., 23 (1897), 1-40. 



42 



CHEMISTRY OF CELLULOSE AND WOOD 



Benedict and Bamberger, 1 who assumed that oak contained 54.1 
per cent 2 of lignin and 2.86 per cent methyl, CH 3 , and that pure 
lignin contained 5.29 per cent methyl. These assumptions are 
incorrect, so that the conclusions of Cieslar must be interpreted 
as only indicating the degree of methylation and not lignification. 



Table 8. — Composition of Alder Wood (Alnus glutinosa) 
(Per cent of dry weight) 



14 years 



Middle 
rings 



Outer 
rings 



70 years 



Heart- 
wood 



Sap- 
wood 



Ash 

a. Ether extract 

Alcohol extract (following a) 

Methyl number 

Methyl alcohol (von Fellenberg) 

Pectin (von Fellenberg) 

, , ... j Calculated as acetic acid 

■ /c f \ l Actual acetic acid 

sis (Schorger) _, . . , 

[ Formic acid 

Nitrogen 

Protein (N X 6.25) 

Furfural 

Pentosan 

Methylpentosan 

Cellulose (Cross and Bevan) 

Cellulose, pentosan-free 

Lignin (Willstatter) 



0.50 
0.71 
3.73 
2.92 
0.231 
2.31 
4.52 
4.20 
0.24 
0.30 
1.88 
14.76 
25.15 


56.22 
39.63 
22.97 



0.48 
0.78 
4.89 
2.83 
0.300 
3.00 
3.81 
3.59 
0.17 
0.26 
1.63 
14.06 
23.98 


58.00 
42.19 
23.93 



0.51 
0.64 



0.20 

0.30 

1.87 

14.04 

23.95 


61.58 
44.45 
22.60 



0.64 
0.78 
0.77 
2.91 
0.165 
1.65 
3.23 
2.89 
0.17 
0.24 
1.50 
13.55 
23.10 


58.35 
44.48 
25.75 



0.53 
1.40 
1.89 
2.85 
0.164 
1.67 
3.43 
3.24 
0.15 
0.29 
1.81 
11.04 
18.85 


59.75 
46.45 
24.27 



He concluded that the variation in lignin content within a species 
is greater than between different species. Spruce has more lignin 
within its natural range than in mild situations outside of it. 
Lignin decreases at the upper limit of tree growth. In spruce the 
lignin decreases from the base of the trunk to the top ; the heart- 
wood is richer in lignin than the sapwood and summer wood 
than spring wood. For equal volumes, rapidly grown wood con- 
tains less lignin than slowly grown wood. Abundant nourish- 
ment and favorable light conditions are conducive to lignin 
formation. Wood is richest in lignin where there is the greatest 
demand for mechanical strength. 

1 R. Benedict and M. Bamberger, Monatsh., 11 (1890), 260; Chem. 
Ztg., 16 (1891), 221. 

2 E. Schulze, Chem. Centr., 28 (1857), 321. 



THE COMPOSITION OF WOOD 



43 



Von Euler, 1 using strong inorganic acids for isolating lignin, 
found that in the trunk of conifers the lignin varied from 25 to 33 
per cent, and in branches even reached 36.5 per cent. Nine 
samples of wood, representing Scotch pine and Norway spruce, 
showed a lignin content varying from 26.80 to 31.35 per cent, the 
average being 28.63 per cent. The amount of lignin in a wood 
depends upon the age, part of the tree from which the sample is 
taken, and growth conditions. Klason 2 found the mean lignin 
content of a spruce 100 years old to be 26.6 per cent, while a tree 
80 years old from the same forest contained 29.5 per cent. In 
the case of abnormal growths, such as "red wood, ' the lignin 
content may reach 37.1 per cent. 

According to Johnsen, 3 cellulose decreases from the base to the 
top of the tree. A balsam fir 10.5 inches in diameter and having 
64 annual rings showed an increase in cellulose from the pith to 
the sap wood . 4 The cellulose was determined by heating the wood 

Cellulose, 
Peb Cent 

A. Pith to fifteenth annual ring 51 . 14 

B. From twentieth to thirty-fifth annual ring 53 . 26 

C. Sapwood (outer 15 annual rings) 54 . 21 

in glycerine-acetic acid mixture, followed by chlorination. The 
sapwood of aspen contained 1.3 per cent more cellulose than the 
heartwood. The effect of abnormal growth is illustrated by the 
following analyses of balsam fir (Table 9). Rapid growth would 
appear to produce less cellulose than slow. 

Table 9. — Effect of Abnormal Growth on Balsam Fir 





"Rotholz," 
per cent 


Rapid 
growth, 
per cent 


Slow 
growth, 
per cent 


Cellulose 

Alcohol extract 


39.42 

1.38 

33.60 

2.74 


50.35 
2.05 

24.44 
2.43 


52.85 
2.85 


Lignin (with 72 per cent H 2 S0 4 ). . . . 
Methyl, CH 3 


24.85 
2.62 



1 A. C. von Euler, Cellulosechemie, 4 (1923), 2, 9. 

2 P. Klason, Cellulosechemie, 4 (1923), 81. 

3 B. Johnsen, Pulp Paper Mag., 15 (1917), 333. 

4 B. Johnsen and R. W. Hovey, J. Soc. Chem. Ind., 37 (1918), 132-137T. 



44 



CHEMISTRY OF CELLULOSE AND WOOD 



< 
o 

3 

§ 8? 

I a 

O xn 

§2 

£ >, 

1 s 

M o 

fc o 

« « 

2 += 

5 2 



&H 


d 


o 


,,_i 




IT) 


r/) 


+J 




3 
03 
(11 




Ph 


rH 




< 
1 




1 

d 




rH 




H 




3 




9 




H 









CO 


HI O 


co 


CO 


O CO 


00 


CO 


CM 


iO 


00 


HI 






CO 00 IO CO 


H IO H IO 


IO 00 


HI CD t^ Tf 




'BUIUI'EQ 


^H O CO HI 


Oi 00 N O 


CD CO 


HI IO O rH 






rH rH rH CM 


rH rH rH CM 


CM rH 


rH rH CM CM 




t> 00 Ol CM 


HI o <N IO 


CN hi 


O CO CM O 






CO lO CM <N 


NINiON 


Ol b- 


Ol CM 00 O 




B^a 


co t+h oo oo 


O O 00 CM 


rH CO 




0) 
03 




rH CM CM CO 


CO CM Hi Hi 


CM CM 


CO CM CM CM 




N 00 H lO 


CO O CM t- 


IO HI 


iO t> O CD 






CO CO rH HI 


rH CO CO CO 


IO HI 


^4 r-t HI IO 


« 


Bqd lV 


HI hi iO CM 


O rH HI CD 


r-i OS 


CN rH CM CO 


a 




t> CO <0 Hi 


lO CO CO CO 


IO IO 


IO © K5 lO 




O -* O 1> 


rH O 00 t> 




CO CN rH HI 




uraso^uad^q^aj^; 


co co co hi 


CM HI ^ HI 


CO CO 


HHNO 






rH rH rH rH 




T-i 1-4 


7-i 1-4 j-i y-{ 




co »o t^ tjh 


CM 1> rH CO 


O O 








oo t> co co 


IO © O 00 


Ol CM 


N 00 H 00 




uisso^uaj 


00 CO Ol t> 


O r)l Co N 


CO CD 


O rH CM rH 

CM CM CM CM 






lO O) ffi M 


00 Ol CD Ol 


t-« iO 


Ol CM CO CO 




uiuSirj 


Ol CO CO CO 


O rH 00 CO 


00 00 


CO CO t^ rH 




CO I- 1> 00 


CO CM CO CO 


rH CM 








« NN M 


CM CM <N CM 


CM CM 


CN CM CM CM 






00 CO CM O 


CO N N N 


00 rH 








m io s hi 


rH IO O HI 


O 00 






8SOpH80 


o co oi co 


oo a oo cs 


CD 00 








iO i<5 ■* lj 


iO »o lO lO 


iO iO 


lO iO lO lO 






O iO CO CO 


00 rH CM CO 


rH CN 






treso^uadjAq^aj^; 


HI CN CO HI 

CN <N CM CM 


IN rH CM rH 
CO CO rH rH 


rH O 


co co 00 






iO O CD l> 


t^ t^ CM 00 










OO CS r, 00 


CO HI 00 o 


1-4 CO 


CO CO CO CN 




u^so^uaj 


O) Ol O O! 


00 00 oo ci 


00 GO 


rH O CM CO 






rH rH CM rH 


rH r-< ^-1 rH 


rH rH 


CM CM <N CM 






O CO CO O 


h too M 


CD Ol 


O * NN 




jAxoq^aj,^ 


i> CO CO CM 


oo oo oo o 


iO t> 


CO HI HI CM 




H< iO «5 iC 


lO lO lO CD 


10 «o 


lO "O IO IO 






CO rH O CO 


CM Ol CO CO 


00 00 


hi 00 10 co 




piOB oijaoy 


CM CO 1> CO 


H 00 W N 


10 


co i> t^ 00 




CO CM CO CM 


CO CM CO CM 


co co 


CM rH CO CM 






N CS CO N 


HI O rH 1> 


rH O 


t> rH 00 HI 






N lO O) o 


b- I> OS IO 




N "5 S rl 




HO^N 1 U9 ° jad T 


rH Ol rH 00 


CO 1> CO t> 


Ol iO 


CD O Ol rH 






CM rH CN rH 


rH r-i T-i rH 


1-1 rH 


rH CM rH IN 




t^ co oo co 


N CO CO 00 


Ol CD 




>> 


Jaq^g 


H HI 00 HI 


CM HI rH IO 


CM CO 


HI 00 00 Ol 


3 
'o 


rH O O O 


O O O O 


O O 


oooo 




rH O CM CO 


oooo h a 


iO iO 






HI HI O HI 


Ol o «o 00 


HI OS 


OS CD rH OS 


aa^'BM ^ojj 


CO CO l> HI 


rH CM CM CM 


CD CN 


rH lO CM CO 




rH HI lO CN 


O) O "5 ifl 

CN IO HI HI 


rH t> 


iO CO hi co 






00 CM CM rH 


Ol O 


O rH t> I> 




J9JT3AY piOQ 


iO CN "O CM 


---- 


Hi CN 


r-l HI r-i CM 






h ONN 


00 Ol CD CO 


»o 


CO O 00 CO 




qsy 


CO CO IO CO 


hi co co co 


HI HI 


CN HI rH CM 




O O O O 


oooo 


O O 


OOOO 






HI IO rH CM 


O Ol Ol 00 


HI O 


CM CO 00 CO 








CO t- t^ 00 


"O OS 


Ol t> CM CO 




ajn^sioj/^; 














U) UJ Tjl N 


Hi CO CO CO 


CO CO 


H^ HI H^ HI 






Ol O rH CN 


t> 00 Ol o 


^4 CM 


i> oo co hi 




afduiBg 


N 00 00 00 


00 00 00 Ol 


Ol Ol 


O O rH rH 

CN CM <N <N 








13 


tj 




T3 


T> 




T3 




T3 


TJ 




00 


ood 

WOO 

ood 
woo 


ood 
woo 
ood 
woo 


"8 § 


ood 
woo 
ood 
woo 




I 


H "£ is "C 


o3 fe u is u 


> u 


birch: 
1 sapw 

1 heart 

2 sapw 
2 heart 




a i ft d 
Jh OS 2 c3 • 


popl 
1 sap 

1 hea 

2 sap 
2 hea 


hick 

1 sap 

2 hea 












"6666 


06666 


god 


£ . . . . 

00000 






3 fc fc a r? 


^zzzz 


M Z Z 


s^Z^ 






£ 






>H 






s 




>H 







THE COMPOSITION OF WOOD 



45 



OS CM OS 00 



b» OS "* t» 



1> rH 

<N 00 


co 

CO 


rH 


co 
os 


-H 

OS 


OS 


IO 




CM 


OS 
iO 


t^ 


OS 

OS 


CO 
CM 


rH 
o 


-fl 


CO lO 

OS O 


10 iH 


CM 
<N 


O 

<N 


»o 


co 


CO 
CM 


<N 


O 


CM 

CM 


CO 
CM 


o 

CM 


o 


CM 
CM 


-r 


N 


CM rH 


i> co 

© CO 


00 


CD 
OS 


05 

o 


CO 

oo 


CO 


CO 

co 


CD 

IO 


OS 
(N 


CD 




co 


-H 


i> 


CM 

CM 


OS CM 

CO CO 


00 b- 

CO CD 


co 


CM 


CD 


CO 


co 
IO 


iO 


rH 

10 


IO 


rH 
IO 




CO 


CD 


Os 
CD 


IO 


O CO 
iO CO 


oo oo 
oo io 


o 

CM 


CD 
OS 


co 


o 
oo 


IO 

CD 


IO 


o> 
o 


O 


r* 


co 




CO 
IO 


oo 

CM 


CM 

co 


rH rH 

CM CO 


o o 


^ 


o 


rt 


rt 


^ 


r 1 


CM 


<N 


CN 


CM 


^ 


H 


rt 


M 


rH M 


t> CM 


00 


IO 


io 

CM 


<* 

co 


OS 

oo 


CO 

co 


CO 


CM 


O 
CD 


co 


IO 

co 


CM 


IO 

OS 


OS 


rH 00 

l-H CO 


rH rH 

IN IN 


CI 


rH 
01 


CO 


10 


>o 


CD 


CO 


t^ 


1> 


i> 


o 


00 


CO 


I> 


O rH 


rH -fi 

co i> 


rt< 


o 

CO 


o 


CD 

o 


CO 


CM 


IO 


rH 


co 
o 


co 


U0 

00 


OS 

co 


r* 


CM 
rH 


CO I> 
t> CO 



CO 00 00 CM 

lOlDHH 


CO O ^ 00 
00 H© H 


iO CO 
CM CM 


CM 00 
rH O 


I> OS CM CM 

Nh O ^ 


os co 

O iO 


OS 00 CO CM 
■* ■* IO KJ 


rH CO O OS 
lO IO "O rH 


rH O 

IO IO 


00 CO 
IO lO 


iO iO ic ■* 

iO iO O IO 


OS rH 

rH * 


O O tH r-i 

OS >0 OS OS 


00 OS -H CO 
CO rH CO CO 


rH O 

i-i o 


iO IO 
1> 00 


rH CM CO CO 

O N H iO 


iO CD 
rH iO 


© rH © © 


rH rH CO CO 


CM rH 




O rH rH rH 


o o 


>0 CM CM 00 
CM 00 b- O 


co t- co oo 

O CO CM 00 


i— i CD 

co »o 


l> OS 

rH CD 


rH OS CM CO 

co n oo n 


00 rH 

o o 



00 CD OS t> 



IO 00 CM "H 
OS rH O CD 


IO rH OS t> 

CO OS OS o 


CO o 

rH CO 


O -H 
rH 00 


o 


O CO OS 
O CM O 


lO 

OS 


CM 


lO CO CO »o 


rH CO TH rH 


TH rH 


rH rH 


lO 


lO IO IO 


lO 


CD 


rH OS h- t- 

rH IO rH OS 


N 00 iO O 
b- rH CO CM 


oo co 

CO rH 


IO CO 
O IO 


N 


rH rH rH 
00 HN 


co 

CO 


00 

CD 


CO CM CM CM 


© O O O 




CM rH 


M 


O rH O 


**< 


O 


rH l-H OS N 
l-H 00 CO CD 


iO OS CO CD 

io io co io 


CO iO 


CM b- 
b- b- 


01 

o 


r-l r-t rfH 
rH b- rH 


CD 


OS 
OS 


rH IO rH CM 

CM (N CM CM 


00 O O CO 


b- OS 


r-< CM 


£ 


i-H CM rH 


£ 


OS 


CO i-i 'O CM 

rH i> co co 


00 N O M 
N 00 00 OS 


CO CM 

rH CO 


O <N 
O CO 




• rH l> 

• rH 00 


CO 


00 


o o o o 


O rH <M b. 


lO CO 






• rH rH 


O 


rH 


rH IO CO O 
rH rH l> CD 


CM OS O OS 
rH OS CO rH 


»o oo 

rH CO 


r-< <N 
rH rH 


CM 

co 


IN CD rH 

CM OS O 


OS 


00 
O 


rH O lO CD 


rH CM CM CO 


IO b- 


CO rH 


CM 


CO CO rH 


N 


b- 


IO CO O CD 
IO CO CM b- 


N O) ON 
b- b- t> CM 


IO b- 

io os 


co oo 

rH 00 


co 


rH CM O 

OS O 00 


CM 

OS 


rH 


CM 1> rH rH 


O CM rH CO 


co io 


CM IN 


CM 


rH CO CM 


^ 


rH 


b- CO b- CN 

IO rH CO rH 


CO O CO >o 
rH CO 00 OS 


CO CM 
CM rH 


oo oo 

CM rH 


CD 


HOON 
N rH IN 




O 

co 


o o o o 


o o o o 


O O 


O O 


O 


O O O 


O 


o 


rH O OS OS 

NKOOl 


CM O N CM 
t> CO O b- 


O N 

OS OS 


b- O 

OS c 


OS 
lO 


OS N 00 
CO O rH 


CO 


CM 



OSCrHCM COrHiOcO M •* 

OrHi-HrH 00000000 OSCS 

CMCMCMCM rH rH l-H rH rH rH 



N00 OJ O 

os os os o 

rH rH rH CM 



-o 

o > 

* t 

p, cS 



T3 

° E 
* t 









O rH rH CM CM 



Cl co 
CM CM 



O O 
fc r? 



O o 



° -S 

£ ^ Ih' 

p, o3 re 

_, _ O 



-0 

o > ° > 

a 



iO CD 

o o 

CM CM 



P .. 6 P 



a cs- 
cs 2 



rH l-H OCMCMCOCO <r> r-i 



zzzzzzz 



O O O o O 



o o o o a 






o o 



46 CHEMISTRY OF CELLULOSE AND WOOD 

The most comprehensive investigation of the composition of 
heartwood and sapwood has been made by Ritter and Fleck 1 
(Table 10). In the softwoods, the extracts with water, ether, 
and hot 1 per cent sodium hydroxide are higher in the heartwood 
than in the sapwood. Cellulose and lignin are correspondingly 
lower in the heartwood, though the lignin of white cedar forms 
an exception. The hardwoods may be divided into two groups, 
one having high extractives in the heartwood, and the other 
having high extractives in the sapwood. Where the extractives 
are high the cellulose is correspondingly low. In the hardwoods, 
with the exception of poplar, the lignin is higher in the heartwood 
than in the sapwood. Acetic acid by hydrolysis is in all cases 
higher in the sapwood than in the heartwood. 

Much attention has been given to the transformation of 
sapwood into heartwood, but the reactions actually taking place 
are far from clear. The heartwood is usually sharply defined 
from the sapwood by its darker color, but in some species it is 
very difficult to distinguish a line of demarcation. The heart- 
wood owes its darker color and higher specific gravity to the 
deposition of resins, gums, tannins, and related bodies that have 
undergone oxidation and other changes. In some cases, copious 
deposition of mineral substances takes place in the cells of the 
heartwood. Red beech under normal conditions does not form 
heartwood even in extreme old age. According to Hartig, 2 
false heartwood is due to pathological effects, but it does not 
differ essentially from true heartwood. 

The substance filling the cells of heartwood and giving it 
increased weight and color was called xylochrome by Hartig. 3 
It was subsequently known as "Kernstoff." He thought 
that heartwood was formed by the transformation of the material 
originally present in the sap in the cells of the sapwood. Forma- 
tion of heartwood in the Amygdalw was assumed by various 
investigators 4 to be due to the deposition of cerasin which was 
formed by gummosis of the inner layers of the cell. 

1 G. J. Ritter and L. C. Fleck, J. Ind. Eng. Chem., 15 (1923), 1055. 

2 R. Hartig, "Das Holz der Rotbuche," Berlin (1888), p. 31. 

3 T. Hartig, cited by J. Gaunersdorfer, Akad. Wiss. Wien, 85, 1 (1882), 9. 

4 A. Wigand, Jahrb. wiss. Botanik, 3 (1863), 115-179; H. Karsten, 
Botan. Zeit., 15 (1857), 313-321; A. Trecul, Compt. rend., 51 (1860) 621. 



THE COMPOSITION OF WOOD 



47 



The observation that heartwood was more resistant to decay 
than sapwood led to the theory that it was formed as a protective 
measure, and that the darkening of living wood following an 
injury was the formation of " wound " or protective wood having 
the properties of heartwood. 1 Bohm 2 mentions that tyloses 
and deposits of gummy substances close the vessels and act as 
protective agents. 

The changes occurring in the living wood of Prunus avium after 
wounding were followed by Temme. 3 The greenish- white 
color of the normal wood became yellow to reddish in 8 to 10 
days. The color change was most intensive in the medullary 
rays. The latter contained numerous small brown grains 
attached to the cell well or surrounding the starch grains. In 
the brownest cells the starch had wholly or partially undergone 
transformation into the brown gum. Similar results were 
obtained with non-amygdalaceous trees, such as Juglans regia, 
Quercus pedunculata, and Pyrus malus. The gum was insoluble 
in water and potassium hydroxide and gave mucic and oxalic acids 
on oxidation with nitric acid. The gums, being colloidal, have 
a low diffusibility and remain at the point of origin. Wounding 
caused an increase in the specific gravity of the sapwood. The 
large amount of heartwood, which lowers the ability of the tree to 
transport water, may explain why the acacias are several weeks 
behind such trees as basswood and birch in their growth in spring. 



Species 



Specific gravity 



Heartwood 



Wound 
wood 



Sapwood 



Quercus 'pedunculata 
Gleditsia triacanthos . 

Prunus avium 

Pyrus malus 

Juglans regia 



1.604 
1.574 
1.677 
1.648 
1.177 



1.130 
0.657 
2.187 
1.523 
1.155 



0.946 
0.202 
1.512 
1.162 
1.100 



1 B. Frank, Ber. botan. Ges., 2 (1884), 330; cf. C. Kraus, Ibid., p. LIII. 

2 J. Bohm, Botan. Zeit., 37 (1879), 229. 

3 F. Temme, Landw. Jahrb., 14 (1885), 465-484. 



48 CHEMISTRY OF CELLULOSE AND WOOD 

Herrmann 1 and Tuzson 2 agree with Hartig that the formation 
of false heartwood is pathological. Irritation of the sapwood 
leads to the formation of tyloses and gums, which retard the 
penetrance of mycelia. Von Tubeuf 3 was of the opinion that 
wound wood had no protective action against fungi, and Will 4 
denied that there is necessarily an analogy between heartwood 
and wound wood. In the latter, the protective agency is a sub- 
stance resembling bassorin which does not occur in heartwood. 

Heartwood is not necessarily protective. The " brown" 
heartwood of English oak (Queries robur) as distinguished from 
normal heartwood is due to a fungus which appears to subsist 
mainly on tannin. 5 

According to Gaunersdorfer, 6 the first step in the formation of 
heartwood consists in filling the wood elements with various 
solid substances, principally starch. The parenchymatous tissues 
are rich in tannic acid, which appears to be an intermediate body 
between starch and the final deposits that form the heartwood. 
Formation of heartwood is due to various tannins, gums, and 
resins which differ with the species of wood. 

The problem of heartwood formation has been studied more 
recently by Munch. 7 The brown deposits are not due to secre- 
tions from the wall of the living cells, but to oxidation of the cell 
contents after death ensues. The same deposits are formed in 
cells after being attacked by fungi, so that they cannot serve as 
a protection against them. A certain amount of moisture in 
addition to oxygen is necessary to produce the brown color. 
Heartwood and protective wood are formed from the same cause. 
Oxygen cannot penetrate the plasma membrane, so that the 
contents of living cells are not colored. The fact that heartwood, 
normal and pathological, is specifically heavier and richer in heart- 
wood substance is due to the occurrence of an extraordinary 

1 E. Herrmann, Z. Forst-u. Jagdwesen, 34 (1902), 596-617. 

2 J. Tuzson, " Anatomische und mykologische Untersuchungen uber die 
Zersetzung und Konservierung des Rotbuchenholzes," Berlin (1905), 89 pp. 

3 K. F. von Tubeuf, Z. Forst-u. Jagdwesen, 21 (1889), 385. 

4 A. Will, "Beitrage zur Kenntnis von Kern-und Wundholz," Diss. Bern 
(1899), 92 pp. 

5 P. Groom, Ann. Botany, 29 (1915), 393-408. 

6 J. Gaunersdorfer, Akad. Wiss. Wien, 85, I (1882), 9-41. 

7 E. Munch, Naturwiss. Z. Forsi-u. Landw., 8 (1910), 533-547; 553-569. 



THE COMPOSITION OF WOOD 49 

influx of food materials into the living cells surrounding dead 
portions of wood, so that the activity of the cambium is increased 
sevenfold. Just how this reasoning applies to heartwood is not 
clear. All cells in wood except those in the thin cambium layer, 
wood parenchyma, and medullary rays are dead. The only 
means of transporting food materials to the interior of the tree is 
through the medullary rays which are always surrounded by dead 
cells. 

The formation of heartwood in "sugi" (Cryptomeria japonica) 
is attributed to the darkening of a substance resembling catechol. 1 
Catechol or a related compound appears to be widely distributed 
in plants. Alkalis promoted coloration while acids and rapid 
seasoning prevented it. In the presence of air and moisture 
ammonia is evolved by decomposition of the protein in the wood, 
thus producing favorable conditions for oxidation of the phenols. 
Tinkler 2 suggests that the reddish-brown color of oak may be due 
to a reaction between a nitrogenous compound and tannin, and 
partly to the presence of phlobaphene. If the rapidity with 
which an alkaline solution of pyrogallol absorbs oxygen and turns 
dark is recalled, this forms an acceptable explanation for the 
darkening of wood. Furthermore, it has been observed that the 
catechin content of the wood of Acacia catechu decreases during 
storage and in the standing tree at certain stages. 3 

The conspicuous black deposit in the heartwood of ebony 
(Diospyros ebenum) has been examined by numerous investiga- 
tors. Molisch 4 supposed that the inner layers of the cell wall 
changed into a gum which subsequently underwent humification. 
The heartwood contained 4.63 per cent of humic acid and 1.30 
per cent of carbon (Humuskohle) . The contention that the " car- 
bon" was formed by reduction 5 and that it actually was carbon 
was disputed by Prael, 6 since it could be rendered colorless by a 
mixture of potassium chlorate and hydrochloric acid. Will 7 

1 M. Fujioka and K. Takahashi, /. Forestry, 19 (1921), 844-866. 

2 C. K. Tinkler, Biochem. J., 15 (1921), 477. 

3 P. Singh, Ind. Forester, 41 (1915), 482-485. 

4 H. Molisch, Akad. Wiss. Wien, 80, I (1879), 54-83. 

5 A. Belohoubek, Botan. Centr., 18 (1884), 293. 

6 E. Prael, " Vergleichende Untersuchungen uber Schutz-und Kernholz," 
Diss. Rostock (1888). 

7 A. Will, Diss. Bern (1899), p. 81. 



50 CHEMISTRY OF CELLULOSE AND WOOD 

supports this view since nitric acid also rendered it colorless. 
Hanausek 1 has found a black substance, phytomelane, in the 
pericarp of a large number of composite. The substance resem- 
bles the black deposit in ebony. Its resistance to oxidizing 
agents, even chromic acid, indicates a high carbon content. 

Ash. — The composition of the ash of representative woods will 
be found in Tables 11 and 12. The ash content of domestic 
woods generally falls within the limits 0.2 to 1.0 per cent. Some 
exotic woods contain 4 to 5 per cent. There is no relation 
between ash content and specific gravity. Balsa, one of the 
lightest of woods, contains 2.0 per cent of ash. 

The ash in wood appears to be either strongly adsorbed or 
combined with organic and inorganic matter in an insoluble 
form. A portion can be extracted with water. Finely powdered 
logwood contained 2.32 per cent of ash in the original state 
and 2.03 per cent after leaching. 2 Schroeder 3 reduced the ash 
content of spruce from 0.232 to 0.183 per cent by leaching with 
water. Sacc 4 digested fir wood with hydrochloric acid and 
succeeded in reducing the ash from 0.55 to 0.07 per cent. Wood 
boiled for a long time with sulphuric acid gave an ash that still 
contained salts of the alkalis. 5 Berthelot 6 found that wood char- 
coal heated with 1 per cent hydrochloric acid at 100° retained 5 to 
6 per cent of its total potassium and 5 per cent of its calcium which 
are present as organic compounds. Insoluble potassium salts 
are present in the leaves of oak but only in traces, if at all, in the 
trunk. 7 

Crystalline deposits, especially calcium oxalate, are of frequent 
occurrence in heartwood. Crystals of calcium oxalate occur in 
Juglans, Hicoria, Diospyros, and all species of Quercus. They 
are particularly abundant in the live oaks. 8 The vessels of 

1 T. F. Hanausek, Akad. Wiss. Wien, 116, I (1907), 3-31; Ber. botan. 
Ges., 29 (1911), 558-562. 

2 F. H. Storer, Bull. Bussey Inst., 2 (1877-1900), 44. 

3 J. Schroeder, Thar, forstl. Jahrb., 24 (1874), 55. 

4 F. Sacc, Ann. chim. phys., [3] 25 (1849), 223. 

5 C. Bischof, J. prakt. Chem., 47 (1849), 196. 

6 M. Berthelot, Compt. rend., 141 (1905), 798. 

7 M. Berthelot, Compt. rend., 142 (1906), 313. 

8 S. J. Record, "Economic Woods of the United States," N. Y. (1912), 
p. 21. 



THE COMPOSITION OF WOOD 



51 



Sideroxylon cinereum, a tropical tree noted for its hardness, 
contains large crystals of calcium oxalate. 1 

Table 11. — Composition of Ash of Georgia Woods 1 



XI 

03 


o 


q 


o 
o 


o 

<5 


o 


6 


o 


6 

CO 


9, 
2 
fa 1 


O 



Carya tomentosa 

Quercus rubra 

Quercus alba 

Quercus obtuailoba 

Pinus mitis 

Cornus florida 

Fraxinus americana. . . 

Castanea vulgaris 

Platanus occidentalis. . . 

Pinus palustris 

Magnolia grandiflora. . 



0.73 


18.93 


3.38 


25.21 


6.66 


7.98 


2.06 0.28 


1.80 


0.25 


1.20 


0.85 


16.41 


3.68 


32.25 


3.58 


7.04 


2.29 


0.68 


0.97 


0.21 


1.38 


0.37 


29.90 


1.94 


21.21 


2.43 


6.72 


4.11 


1.00 


3.20 


0.50 


3.87 


1.09 


15.46 


7.22 


32.67 


4.84 


6.38 


0.38 


0.77 


1.00 


0.37 


0.65 


0.35 


12.97 


1.18 


43.31 


2.11 


2.75 


0.86 


0.67 


2.35 


0.18 


0.74 


0.95 


20.03 


7.65 


27.81 


4.92 


1.02 


2.72 


0.74 


2.00 


0.12 


0.40 


0.43 


34.74 


0.94 


17.68 


0.45 


2.69 


8.5S 


0.26 


7.05 


2.92 


1.14 


0.22 


13.33 


7.67 


36.43 


1.56 


5.00 


2.68 


0.88 


1.17 


2.65 


1.19 


0.99 


18.24 


5.92 


24.98 


0.49 


9.65 


5.73 


0.45 


9.66 


4.13 


2.14 


0.49 


10.34 


2.34 


37.24 


4.21 


2.65 


4.32 


0.21 


3.41 


2.76 


1.11 


0.60 


11.87 


2.52 


23.64 


4.89 


5.31 


3.46 0.23 


7.32 


1.60 


17.22 



32.44 
31.85 
25.16 
28.86 
33.26 
28.16 
23.65 
24.82 
19.01 
31.47 
22.16 



1 H. C. White, Ga. Agr. Exp. Bull., 2 (1889), 17-26; 3, 50-53. 

2 Based on wood with 10 per cent moisture. 

Table 12. — Composition of Ash of Heartwood and Sapwood 1 
(Results based on 10,000 parts of dry wood) 



O 

bJO 



Larch : 

Sapwood . . 

Heartwood 
Pine: 

Sapwood . . 

Heartwood 
Spruce : 

Sapwood. . 

Heartwood 
Oak: 

Sapwood. . 

Heartwood 
Beech : 

Sapwood. . 

Heartwood 



5.32 

2.97 

5.40 
2.30 

9.80 
5.93 

19.56 
6.71 

19.24 
15.47 



1.28 
0.15 

1.37 
0.13 

2.87 
0.20 

5.22 
0.43 

2.15 
0.60 



3.33 
1.94 

2.09 
2.41 

1.46 
1.96 

2.63 

0.44 

6.77 
5.11 



1.10 
0.93 

1.19 
0.83 

1.54 
1.71 

1.46 
0.51 

1.65 

0.82 



1.27 
0.55 

0.98 
0.68 



1.08 
0.25 

0.39 
0.52 



1.110.92 
. 86 . 20 



2.90 
1.98 

1.93 
1.55 



0.56 
0.88 

0.59 

0.82 



0.90 
0.59 

0.87 
0.47 

0.38 
0.65 

1.13 
0.24 

0.53 

0.17 



1 W. Daube, Forstliche Blatter, 20 (1883), 177. 

2 Per cent of dry wood. 



1 H. Molisch, Akad. Wiss, Wien, 80, I (1879), 75. 



52 



CHEMISTRY OF CELLULOSE AND WOOD 



Baker 1 examined 22 orders of Australian trees and found 
crystals of calcium oxalate in 15 of them. Usually only one 
crystal occurs in a cell. The "grit" of Eucalyptus pilularis and 
Tristania conferta is due to silica. Molisch 2 found copious 
deposits of calcium carbonate in the heartwood of Ulmus, Celtis, 
Fagus, and other genera. The pith of Populus alba contained 
38.9 per cent ash. The principal deposition of crystalline calcium 
carbonate takes place in the heartwood, owing to its lower con- 
ductivity; as the solution of calcium bicarbonate slowly rises carbon 
dioxide escapes, due to the increased temperature, calcium car- 
bonate being deposited. The heartwood of red carrabeen 
(Geissois benthami) contains deposits of calcium carbonate which 
in some cases form 18 per cent of the weight of the wood. 3 
Cracks in the wood of Hieronyma alchorneoides from Trinidad 
contained a deposit, of which 85.8 per cent was calcium carbonate. 4 

In general, the sapwood contains more ash than the heartwood, 
which is to be expected, since it contains the tissues engaged in 
the transportation of water and foodstuffs. Daube 5 (Table 12) 
found the most ash in the sapwood where potassium, phosphorus, 
and sulphur were most abundant. Excessive deposition of 
calcium in the heartwood may reverse the ash content. Zimmer- 
mann 6 found the distribution of ash in a beech (Fagus silvatica) 94 
years old to be as follows : 



Annual rings. 

Ash, per cent 
CaC0 3 



1-15 

1.162 
0.579 



15-25 

0.827 
0.251 



25-35 

0.645 

Trace 



35-45 

0.612 

Trace 



45-60 



0.555 



60-83 



0.458 



83-94 

(sapwood) 

0.205 



The ash is most abundant in the roots where the transportation 
of water begins and in the twigs where it ends. In the trunk the 



1 R. T. Baker, J. Proc. Roy. Soc. N. S. Wales, 61 (1917), 435-444. 

2 H. Molisch, Akad. Wiss. Wien., 84, I (1881), 7-28. 

3 T. Steel, Proc. Linnean Soc. N. S. Wales, 46 (1921), 489. 

4 J. H. Hart, Ann. Botany, 1 (1887), 361-362. 

5 W. Daube, I.e. 

6 H. Zimmermann, Z. angew. Chem., 6 (1893), 426-428. 



THE COMPOSITION OF WOOD 53 

ash decreases from the base upwards. In general, there is little 
variation in a given portion of the wood throughout the year. 1 

The role of the inorganic constituents is not definitely known. 
In spite of the composition of the soil the constituents are usually 
taken up selectively. The ash of pines growing in a sandy soil is 
Low in silica and high in lime (Table 11). The potash content 
varies from 10 to 35 per cent. In general, the hardwoods are 
highest in potash but there are numerous exceptions. According 
to Sieber, 2 during the formation of buds, calcium decreases in the 
cambium and potassium increases. During the period of assimila- 
tion, the calcium increases and by summer may double in amount. 
Lime may form 75 per cent of the total ash. The following 
values are given by Wolff: 3 

CaO CaO 

Tilia grandifolia 75 . 92 Ulmus campestris 77.31 

Robinia pseudacacia 58 . 30 Fagus silvatica 60 . 25 

Fraxinus excelsior 62 . 14 Quercus pedunculata 76 . 27 

Populus tremula 66 . 50 Sorbus aucuparia 76 . 13 

Abies pectinata 10 . 17 Picea excelsa 29 . 41 

Ebony (Diospyros ebenus) contains 3.92 per cent ash, of which 
90 per cent is calcium carbonate. 4 

Magnesium, manganese, iron, aluminum, chlorine, sulphur, 
phosphorus, and silicon usually form only a minor portion of the 
ash. Manganese appears to be an essential constituent and in 
some cases occurs in large amounts. Schroeder 5 found 22.47 per 
cent Mn 3 4 in the ash of Picea excelsa and as much as 40 per cent 
in that of Abies pectinata. Guerin 6 extracted from various woods 
a nitrogenous product high in manganese. According to 
Teicher, 7 plants growing in wet situations contain more 
manganese than those growing on higher ground; conifers con- 

1 The ash content of trees has been extensively investigated from all 
angles. Consult: E. Wolff, "Aschen-Analysen," Berlin, Part I (1871), 
194 pp. and Part II (1880), 170 pp.; F. Czapek, "Biochemie der Pflanzen," 
Jena, 2 (1920), 400-414. 

2 F. W. Sieber, Verhandl. phys.-med. Ges. Wurzburg, 41 (1912), 11. 

3 E. Wolff, I.e. 

4 H. Molisch, Akad. Wiss. Wien, 80, I (1879), 70. 

5 J. Schroeder, Thar, forstl. Jahrb., 24 (1874), 275. 

6 G. Guerin, Compt. rend., 125 (1897), 311. 

7 Teicher, Cellulosechemie, 2 (1921), 64. 



54 CHEMISTRY OF CELLULOSE AND WOOD 

tain more manganese than the hardwoods. The ash of teak wood 
(Tectona granda) consists largely of calcium phosphate. 1 
Frankforter 2 reports the presence of metallic copper in an oak 
tree at Minneapolis. In the last six to seven annual rings the 
copper could be seen with the naked eye. 

Aluminum is usually present in very small amounts. The 
" silky oak" (Orites excelsa) is exceptional; its ash (0.7 per cent) 
contains 40 to 80 per cent of alumina. 3 The metal occurs 
in the wood as basic aluminum succinate. A trace of cobalt was 
found in the ash from one specimen. 

Pectin. — The nomenclature of pectin and its related 
compounds is confusing and greatly in need of standardization. 
According to von Fellenberg, 4 there are three compounds, pectose 
pectin, and pectic acid, that must be considered. Pectose or 
protopectin occurs in unripe fruits. It is converted into soluble 
pectin by ripening, and by boiling with water or dilute acids. 
Alkalis convert it into a salt of pectic acid, from which pectic acid 
is obtained on acidifying. Schryver and Haynes 5 claim that 
there is but a single acidic substance, pectinogen, to be obtained 
from all sources. Weak alkalis convert it to pectin from which 
gels can be obtained. Addition of lime water or calcium chloride 
precipitates the pectin quantitatively. The actual existence of 
protopectin is doubted by Tutin, 6 since he found that if the cell 
structure of green apples was sufficiently destroyed all the pectin 
could be removed with cold water. Sucharipa, 7 on the other 
hand, furnishes evidence that in the rind of Citrus the protopectin 
is combined with cellulose. 

It is highly probable that various pectins occur in nature. 
Von Fellenberg 8 was the first to observe that pectin gives methyl 
alcohol on treatment with dilute alkali. He considered pectin to 
be the methyl ester of pectic acid. While fruit pectins give 
about 10 per cent of methyl alcohol, that from pea pods contains 

1 G. Thoms, Landw. Vers.-StaL, 23 (1879), 413. 

2 G. B. Frankforter, Chem. News, 79 (1899), 44. 

3 H. G. Smith, J. Proc. Roy. Soc. N. S. Wales, 37 (1903), 107-120. 

4 T. von Fellenberg, Mitt. Lebensm. Hyg., 5 (1914), 225. 

s S. B. Schryver and D. Haynes, Biochem. J., 10 (1916), 539-547. 

6 F. Tutin, Biochem. J., 17 (1923), 510-514. 

7 R. Sucharipa, J. Am. Chem. Soc., 46 (1924), 145-156. 

8 T. von Fellenberg, Mitt. Lebensm. Hyg., 4 (1913), 122, 273. 



THE COMPOSITION OF WOOD 55 

only 1.9 per cent. 1 In flax, the methoxyl content of the pectin 
varies with the part of the plant from which the pectin is obtained 
and the treatment which it has received. 2 

Tutin 3 found that acetone, also, was obtained by treating 
pectin with 0.1A r sodium hydroxide at room temperature. The 
ratio of acetone to methyl alcohol was 1:2. Both compounds 
are believed to be present as ester groupings, the acetone in its 
enolic form, — CH 3 .C (OH) : CH 2 . Pectin is probably the dimethyl- 
isopropenyl ester of pectic acid. 

It has long been known that pectin contains araban and gives 
mucic acid on oxidation. The most illuminating work on the 
constitution of pectin is the paper of Ehrlich. 4 Crude pectin 
from beets consists of the calcium and magnesium salts of pectic 
acid. Pure pectic acid contains about 9 per cent of methyoxyl 
and is practically free from pentoses, though it gives furfural on 
distillation with acids, and shows the pentose reaction with 
orcinol and resorcinol. Limited hydrolysis with dilute oxalic 
acid gives d-galactose-galacturonic acid, Ci 2 H 2 oOi 2 . Sodium 
hydroxide produces d-tetragalacturonic acid, C 24 H 3 40 25 , which 
shows the color reactions for pentoses and glucuronic acid. 
Pectin in nature is the calcium-magnesium salt of an anhydro- 
arabino-galactose-m ethoxyltetragalacturonic acid . 

Nanji 5 recognizes only pectinogen and pectic acid which is 
formed by de-esterification. The basal molecule consists of one 
anhydroarabinose group, one anhydrogalactose group, and four 
galacturonic acid groups. Methylpentoses are absent. Iron is 
a more important constituent of the pectinogen than calcium or 
magnesium. 

Only small amounts of pectin exist in wood. Von Fellenberg 6 
obtained about 0.05 per cent of methyl alcohol (0.5 per cent 
pectin) by the alkaline hydrolysis of fir wood. Pectic acid could 
not be detected nor could pectin be obtained by heating the wood 

1 R. G. W. Farnell, Int. Sugar J., 25 (1923), 248-251. 

2 E. Correns, Faserforschung, 1 (1921), 229-240. 

3 F. Tutin, Biochem. J., 15 (1921), 494-497. 

4 F. Ehrlich, Chem. Ztg., 41 (1917), 197-200. 

5 D. R. Nanji, F. J. Paton, and A. R. Ling, J. Soc. Chem. Ind., 44 (1925), 
253-258T. 

6 T. von Fellenberg, Biochem. Z., 85 (1918), 82; Mitt. Lebensm. Hyg., 
8 (1917), 1-29. 



56 CHEMISTRY OF CELLULOSE AND WOOD 

with water or 1 per cent tartaric acid at 120°. The derivative 
yielding methyl alcohol is, accordingly, not identical with fruit 
pectin. There is more pectin in young wood than in old and more 
in hardwoods than in conifers. The wood of an ash seedling 
contained 1.6 per cent pectin. Schwalbe and Becker, 1 using von 
Fellenberg's method, found 3.0 per cent of pectin in alder wood. 
Nanji 2 determines pectin by taking advantage of the property 
of uronic acids of forming carbon dioxide when distilled with 12 
per cent hydrochloric acid. By this method the cereal straws 
were found to contain 6.8 to 9.5 per cent of pectin. In wood 
and straw, only a portion of the pectin is soluble in oxalic acid 
solution. O'Dwyer 3 has recently isolated from beech wood a 
small amount of a pectic substance. 

Our chief interest in pectin lies in its relation to lignin and the 
middle lamella which botanists believe consists of modified 
pectin. Treub 4 suggested that in cell division the protoplasmic 
cell plate splits to form the plasma membrane of the two daughter 
cells. Between these membranes is secreted a substance which 
becomes the primary layer or middle lamella. This view has 
been confirmed by Timberlake 5 and Allen. 6 

The generally accepted view as to the chemical composition of 
the newly formed cell walls are due principally to Mangin. 7 He 
found that the middle lamella consisted of pectose, on which was 
deposited a secondary layer of cellulose. The various layers 
rarely exist in a pure form and soon undergo profound modifica- 
tion. The pectose of the primary layer changes to insoluble 
calcium pectate, while that in the secondary and tertiary layers is 
transformed into lignin; reactions for cellulose are then no longer 
obtained. 

It was found by von Fellenberg 8 that the youngest growth of 
ash was rich in pectin but poor in lignin. His passing suggestion 

1 C. G. Schwalbe and E. Becker, Z. angew. Chem., 33 (1920), 14. 

2 D. R. Nanji et al., I. c. 

3 M. H. O'Dwyer, Biochem. J., 19 (1925), 694-696. 

4 M. Treub, "Quelques recherches sur la role du noyau dans la division 
des cellules vegetales," Amsterdam (1878). 

5 H. G. Timberlake, Botan. Gaz., 30 (1900), 73-99; 154-170. 

6 C. E. Allen, Botan. Gaz., 32 (1901), 1-34. 

7 L. Mangin, Compt. rend., 107 (1888), 144-146. 

8 T. von Fellenberg, Biochem. Z., 85 (1918), 88, 94. 



THE COMPOSITION OF WOOD 57 

that the young growth changes to wood poor in pectin and rich 
in lignin is worthy of careful consideration ; both pectin and lignin 
are high in carbon and methyl groups and both are associated 
with pentosans and galactan. 

Both pectose and pectic acid are associated with the cellulose. 1 
When sections were treated with acid-alcohol mixture, then 
stained with a pectic acid stain such as methylene blue, Bismark 
brown, or other basic dye, the middle lamella stained strongly 
while the pectin in the cellulose layers was much less strongly 
colored. Mangin inferred that a neutral pectin occurred with the 
cellulose and that there was a salt of pectic acid in the middle 
lamella. Thin sections were treated with acid-alcohol mixture, 
washed with water, and immersed in an alkaline solution. 2 After 
a short time the mass was agitated and filtered. Acidification 
of the filtrate produced a gelatinous mass of pectic acid. 

According to Devaux, 3 the middle lamella does not consist of 
calcium pectate but of pectose. The latter is hydrolyzed by an 
alcoholic solution of hydrochloric acid and also by heating 
with water for a few minutes. The soluble pectin thus produced 
is transformed into pectic acid by alkalis. The cell wall contains 
several pectoses, which under the action of reagents show different 
resistances to transformation into pectin. 

Devaux 4 observed that pectic compounds reacted with salts 
of the heavy metals to give a salt of pectic acid, while lignified 
tissues were scarcely affected until after treatment with hypo- 
chlorite solutions. A section is plunged into a solution of a 
ferric salt, then washed with water and finally with a 2 per cent 
solution of acetic acid. The iron is invisible, but is brought out 
by immersion in a solution of potassium ferrocyanide. The blue 
color produced is accentuated by a drop of hydrochloric or nitric 
acid. The coloration is especially strong for the soft tissues. 
After washing with dilute acid, the lignified tissues are so feebly 
colored that the wood can be stained with safranin, thus produc- 
ing double colorations. 

1 L. Mangin, Compt. rend., 109 (1889), 579-582. 

2 L. Mangin, Compt. rend., 110 (1890), 295-297. 

3 H. Devaux, Mem. soc. sci. phys. nai. Bordeaux, [6] III (1903), 89. 

4 H. Devaux, Act. Soc. Linn. Bordeaux, [VI] 6 (1901), XXXIII, LVIII, 
and LXXXII. 



58 CHEMISTRY OF CELLULOSE AND WOOD 

Mangin 1 found in ruthenium red, Ru 2 (OH)2Cl4.7NH 3 .3H 2 0, 
what he considered to be a specific stain for pectin. Lignin is 
not colored strongly until after treatment with alkalis or 
hypochlorites. In addition to pectin, ruthenium red stains 
gums, mucilages, nitrogenous compounds, and the cuticle of 
cotton, but not the cutinized membranes of leaves and twigs. 
The color with pectins is unaffected by glycerine and alcohol. 
An aqueous solution of the stain of a concentration of 0.005 per 
cent is satisfactory. 

Stains seldom furnish satisfactory information as to chemical 
composition. It may well be doubted if ruthenium red is 
specific and capable of distinguishing between lignin and pectin. 
Phloroglucinol-hydrochloric acid, which is a reagent for lignin, 
stains the middle lamella of lignified elements, e.g., Tilia, Quercus, 
and Pinus, more deeply than the secondary membranes. 2 What- 
ever may be the relation between pectin, lignin, and the middle 
lamella from a botanical standpoint, there is at present no strik- 
ing chemical analogy apparent between lignin and pectin. The 
galacturonic acid groups of the pectin are so profoundly modified 
in lignin that they are no longer capable of giving mucic acid on 
oxidation. The methyl groups having an ester linkage in pectin 
have an ether linkage in lignin. These changes take place with 
such rapidity in the cambium, which at best forms only a small 
portion of the wood, that it is impossible to follow them by 
ordinary chemical methods. The solution will probably have 
to await further development in the technique of microchemical 
analysis. 

Sachsse 3 looked upon lignin as a transformation product of 
cellulose. This hypothesis has been supported by others. 4 
According to Konig and Rump, 5 lignin is formed by the addition 
of alkyl groups to cellulose. The changes must be more pro- 

1 L. Mangin, Compt. rend., 116 (1893), 653-656. 

2 F. C. von Faber, Ber. botan. Ges., 22 (1904), 180. 

3 R. Sachsse, "Die Chemie und Physiologie der Farbstoffe, . . . " 
Leipzig (1877), p. 133. 

4 C. F. Cross and E. J. Bevan, "Cellulose" (1895), p. 179; A. G. Green, 
Z. Farben-u. Textilchem., 3 (1904), 97; K. Fromherz, Z. physiol. Chem., 
60 (1907), 210. 

5 J. Konig and E. Rump, "Chemie . . . Pflanzen-Zellmembran " 
(1914), p. 83. 



THE COMPOSITION OF WOOD 59 

found, since trimethylcellulose contains only 53 per cent of carbon 
while they report 68.75 per cent for wood lignin. 

There is also the hypothesis that lignin is derived from 
pentoses. 1 Klason 2 believes that lignin, like the tannins to 
which it is closely related, may exist in wood as a glucoside. 
The parent substance, dihydroxycinnamyl alcohol, may be 
formed from the pentoses as follows : 

2C 5 H 10 O 5 = C 9 H 10 O 3 + 5H 2 + C0 2 . 
Pentose Dihydroxycinnamyl alcohol 

Methylation of the phenolic groups may take place through 
formaldehyde : 

R.OH + CH 2 = R.OCH 3 + 0. 

The oxygen liberated changes R.H into R.OH and oxidizes allyl 
alcohol complexes into the corresponding aldehyde and acid 
complexes occurring in lignin. A similar derivation is assumed 
by von Euler. 3 Coniferyl alcohol, formed from pentoses, is the 
source of related alcohols and aldehydes, which condense to form 
a-lignin, and related acids which form /3-lignin and tannic acids. 

Strupp 4 has recently elaborated a derivation of lignin from the 
cyclic sugars, such as pinite and inosite. 

Starch. — Starch in small amounts appears to be found locally 
throughout the entire trunk of the tree. According to Sinnett, 5, 
it is most abundant in winter in the thick-walled, well-lignified 
cells remote from the centers of formation. 

During the dormant period starch is stored in the living cells of 
the pith, wood parenchyma, and medullary rays. 6 When 
vegetation begins it appears in the bark tissues. With the 
formation of leaves the starch disappears from the bark, wood 
parenchyma, medullary rays, and pith in the order given. A 
portion may never be used during the growth of the tree, but is 

l K. G. Jonas, Z. angew. Chem., 34 (1921), 290; B. Rassow and A. 

ZSCHENDERLEIN, Ibid., p. 204. 

2 P. Klason, Arkiv Kemi, Mineral Geol, 6, No. 15 (1917), 20; cf. Ber., 
55 (1922), 454. 

3 A. C. von Euler, Cellulosechemie, 3 (1922), 3; cf. C. A., 15 (1921), 2470. 

4 E. Strupp, Cellulosechemie, 5 (1924), 6. 

5 E. W. Sinnett, Botan. Gaz., 66 (1918), 162. 

6 W. A. Price, Ohio J. Sci., 16 (1916), 356. 



60 CHEMISTRY OF CELLULOSE AND WOOD 

occluded in cells changed into heartwood, where it may remain 
indefinitely. This starch is considered as rendering the wood 
susceptible to decay. It has never been shown that starch influ- 
ences durability; it is a popular belief, however, that wood felled 
in summer is best, since it contains the least starch at this season. 
Experiments conducted in Germany showed that the month of 
felling was without influence on durability. 1 Western larch is a 
comparatively durable wood, although it contains a large amount 
of an analogous carbohydrate, galactan. 

Hartig 2 believed that the greater part of the starch stored in 
autumn was held against the fruiting season, during which it 
disappeared rapidly and completely. Spring growth exhausted 
only the starch in the one or two youngest annual rings. 

The data available on the starch content of woods are meager, 
and generally misleading, owing to faulty methods of analysis. 
Methods based on acid hydrolysis 3 are in some cases grossly 
incorrect, as pentosans and other hemicelluloses are hydrolyzed 
simultaneously with the starch. Du Sablon 4 used acid hydrolysis 
for determining the variation in starch content of various fruit 
and forest trees throughout the year. The trunk wood of chest- 
nut, e.g., is credited with 17.6 to 24.2 per cent of starch. 

The suggested correction for pentoses 5 does not provide for 
hexosan hemicelluloses. 

The use of diastase 6 or takadiastase 7 likewise does not give 
accurate results. 8 

The best data available indicate that the starch content of 
wood does not exceed 3 to 4 per cent. In mature forest wood it is 
probably considerably lower. The wood of the pear tree con- 

1 Wagner's Jahresber., 29 (1883), 1200. 

2 R. Hartig, Botan. Zeit., 46 (1888), 837-842. 

3 "Methods of Analysis A. O. A. C." Washington (1920), p. 95; von 
Wissel, Landw. Jahrb., 53 (1919), 617; cf. A. Manaresi and M. Tonegutti, 
Staz. sper. agrar. ital., 43 (1910), 705. 9 

4 L. dxj Sablon, Compt. rend., 135 (1902), 866; 140 (1905), 1608; Rev. 
gen. botan., 16 (1904), 341; 18 (1906), 5, 82. 

5 S. Weiser and A. Zaitschek, Arch. Physiol, 93 (1903), 98. 

6 H. W. Wiley and W. H. Krug, /. Am. Chem. Soc, 20 (1898), 253, 266. 

7 W. A. Davis and A. J. Daish, /. Agr. Sci., 6 (1914), 152. 

8 F. E. Denny, J". Assoc. Official Agr. Chem., 6 (1922), 175; W. E. Tot- 
tingham and F. Gerhardt, J. Ind. Eng. Chem., 16 (1924), 139. 



THE COMPOSITION OF WOOD 



61 



tained 3.07 per cent of starch. 1 Tottingham and Gerhardt 2 
found that digestion with saliva was the most reliable method. 
Young branches of apple and plum trees contained 3.3 and 3.9 
per cent of starch respectively. Beckmann 3 determined the 
amount of starch in trees felled in autumn and found : 

Starch, 
Species Per Cent 

Maple 2.65 

Birch 0.95 

Alder 1.54 

Elm 5.90 

Sweet chestnut 2 . 65 

The starch was determined by exhausting the finely powdered 
wood with water at 108 to 110° for 8 hours and subsequently 
heating the filtrate (100 cubic centimeters) with 10 cubic centi- 
meters of 25 per cent hydrochloric acid for 3 hours on the water 
bath. The reducing value of the sugar formed was calculated 
as starch. Wood of birch, maple, alder, and elm collected in the 
spring showed no perceptible bluing with iodine-potassium iodide 
solution, but aspen and willow gave a strong starch coloration. 

Storer, 4 using diastase, found the following amounts of starch : 



Inner 

wood, 

per cent 



Outer 

wood, 

per cent 



Bark, 
per cent 



Gray birch felled in May 

Gray birch felled in July 

Gray birch felled in October. . 
Sugar maple felled in October 
Alder felled in January 



4.93 
3.83 
3.75 
1.94 



5.42 
3.87 
3.51 
2.43 
3.04 



7.67 
7.52 
4.24 
5.97 



The starch content of the twigs of various trees has been 
determined microscopically 5 and polarimetrically after hydrolysis 
by heating with 1 per cent hydrochloric acid on the water bath 
for 15 minutes. 6 

1 A. Manaresi and M. Tonegutti, I.e., p. 714. 

2 W. E. Tottingham and F. Gerhardt, I.e. 

3 E. Beckmann, Chem. Centr., II (1915), 1208. 

4 F. H. Storer, Bull. Bussey Inst., 2 (1897), 389. 

5 R. Lucks, Landw. Jahrb., 53 (1919), 586-615. 

6 Von Wissel, Landw. Jahrb., 53 (1919), 617-625. 



62 



CHEMISTRY OF CELLULOSE AND WOOD 



Sugar. — Little information is available on the amount of sugar 
present in wood. Fischer 1 has made qualitative tests with 
Fehling's solution. The sap of the sugar maple contains an 
average of 3 per cent of sucrose. It has been calculated that the 
average sugar maple contains 135 gallons of sap or 35 pounds of 
sugar. 2 

The "sand" obtained in the manufacture of maple syrup 
contains 65 to 80 per cent of calcium malate and 6 to 18.5 per 
cent of silica. 3 In addition to malic acid, small amounts of 
d-tartaric acid and tricarballylic acid have been found. 4 It has 
been observed that in sunlight the crushed tissues of maple 
shoots can partially convert malic acid into sugar. 5 

Percentage of Available Carbohydrate Food in Maple Wood 



Determination 


December, 
January, 

and 
February 


March 
and 
April 


May, June, 

July, and 

August 


September, 
October, 

and 
November 


Outer base wood: 

Water 


35.01 
0.94 
1.40 

0.78 

30.28 
0.49 
0.65 
0.43 

33.85 
1.12 
1.79 
1.09 


32.86 
0.80 
0.53 
0.96 

29.64 
0.52 
0.33 
0.49 

31.36 
1.04 
0.67 
1.33 


30.30 
0.36 
0.13 
1.01 

26.87 
0.18 
0.09 
0.62 

32.46 
0.41 
0.23 
1.23 


29.80 


Sucrose 


0.72 


Reducing sugars 

Starch 


0.41 
1.43 


Inner base wood: 

Water 


29.05 


Sucrose 


0.41 


Reducing sugars 

Starch 


0.26 
0.51 


Top wood (outer and inner) : 
Water 


32.14 


Sucrose 


0.83 


Reducing sugars 

Starch 


0.59 
1.63 







1 A. Fischer, Jahrb. wiss. Botanik., 22 (1891), 73-160. 

2 Vermont Agr. Exp. Stat. Bull, 105 (1904). 

3 J. F. Snell and A. G. Lochhead, J. Ind. Eng. Chem., 6 (1914), 301- 
302; W. H. Warren, J. Am. Chem. Soc, 33 (1911), 1205-1211. 

4 E. O. von Lippmann, Ber., 47 (1914), 3094-3095. 

5 W. R. Bloor, J. Am. Chem. Soc, 34 (1912), 534-539. 



THE COMPOSITION OF WOOD 63 

The amount of starch and sugars present in the wood of the 
sugar maple have been determined by Jones. 1 (See above Table.) 
The sugar content is highest in winter and lowest in summer. 
The sugar is practically limited to the sapwood, especially the 
outer layers. The per cent of sugar present based on the entire 
wood of the trunk would be exceedingly small. 

Nitrogen. — The percentages given in the early analyses of 
wood are usually much too high. Most woods contain 0.1 to 
0.3 per cent nitrogen (Tables 3 and 4). Tinkler 2 found 0.331 
and 0.404 per cent nitrogen in various kinds of oak. Storer 3 
found 0.10 to 0.156 per cent in fresh logwood and 0.141 to 0.164 
per cent after leaching. The outer wood of a gray birch felled in 
October contained 0.52 per cent. According to Hartig, 4 nitro- 
gen decreases during the fruiting period. Beech wood contained 
0.098 to 0.392 per cent nitrogen and this decreased to 0.01 per 
cent during a year of heavy fruiting. 

Konig 5 states that wood loses nitrogen by soaking in a solution 
of copper sulphate. 

The nature of the nucleic acids in the protein of wood does not 
appear to have been investigated. Guerin 6 digested wood with 1 
per cent potassium hydroxide for several days. The extract 
contained a nitrogenous compound of the following composition : 

Per Cent 

C 52.762 

H 5.040 

S 0.666 

P 4.600 

Mn 0.402 

The high manganese content is of interest. It appears to be 
combined in this manner in the woody tissues of all plants. 

Nitrogen also occurs in wood in the form of alkaloids and other 
compounds. The wood of Celtis reticulosa, a large tree growing in 

1 C. H. Jones, J. Assoc. Official Agr. Chem., 2 (1916), 103-111. 

2 C. K. Tinkler, Biochem. J., 15 (1921), 482. 

3 F. H. Storer, Bull. Bussey Inst., 2 (1877-1900), 28. 

4 R. Hartig, Botan. Zeit., 46 (1888), 837. 

5 R. Konig, Chem. Centr., 32 (1861), 588. 

6 G. Guerin, Compt. rend., 125 (1897), 311-312. 



64 CHEMISTRY OF CELLULOSE AND WOOD 

the East Indies, contains indol and skatol; their origin is probably 
pathological. 1 

Secondary Constituents. — The secondary compounds occurring 
in wood are many and varied; they include gums, oleoresins, 
volatile oils, tannins, coloring matters, etc., some of which are of 
great economic importance. 

A few woods are rich in tannins, while all appear to contain 
small amounts of these or related bodies. The tannins may be 
roughly divided into two groups, catechol and pyrogallol; the 
members of the former give a greenish-black coloration with 
ferric alum, and the latter a bluish-black. Moureau and 
Dufraisse 2 have observed that some phenols possess to a surprising 
degree the property of preventing oxidation ; thus one molecule of 
hydroquinol will protect 40,000 molecules of acrylaldehyde from 
oxidation. Owing to the wide distribution of phenols in the 
vegetable kindgom, it is suggested that their function is to 
protect the plants from atmospheric oxidation. 

Examination of the secondary constituents is in some cases an 
excellent means for distinguishing between species or varieties 
where the botanical differences are questionable. Schorger 3 
examined the volatile oils from the oleoresins of Pinus ponderosa 
and P. ponderosa scopulorum and found that they were optically 
and chemically different. Henry and Flood 4 found differences 
in the composition of the oils from the needles of Douglas fir, 
Pseudotsuga taxifolia, and its variety P. taxifolia glauca. 

Identification of woods by the character of the various extracts 
obtainable from them is capable of important development. 
Jauffret 5 has examined a large number of colored tropical woods 
and tabulated the following data : colorations obtained by treat- 
ing the water and alcohol extracts with normal solutions of ferric 
chloride, sulphuric acid, sodium hydroxide, calcium hypochlorite, 
and sodium bisulphite; spectroscopic characteristics of the 

1 W. R. Dunstan, Proc. Roy. Soc, 46 (1889), 211; C. A. Herter, J. 
Biol. Chem., 5 (1909), 489-492. 

2 C. Moureau and C. Dufraisse, Compt. rend., 174 (1922), 258-264. 

3 A. W. Schorger, Proc. Soc. Am. Foresters, 11 (1916), 33-39. 

4 A. Henry and M. G. Flood, Proc. Roy. Irish Acad., 35B (1920), 67-92. 

5 A. Jauffret, "La determination des bois exotiques colores d'apres 
leurs caracteres chimiques et spectroscopiques," Ann. Mus. Colonial de 
Marseille, [3] 8 (1920), 1-171. 



THE COMPOSITION OF WOOD 



65 



alcoholic (+NH 4 OH) and aqueous (+NaOH) extracts. In 
most cases the chemical and spectroscopic properties of each 
species were distinct. 

Kanehira 1 has studied the fluorescence of the aqueous extracts 
of numerous species of American woods as an aid to identification. 

Romanis 2 found that the resinous substances in fresh teak wood 
undergo a decided change with age. Advantage has been taken 
of this phenomenon to determine the degree of seasoning. 3 
Steam distillation of fresh teak yields an oil, while crystals only 
are obtained from fully seasoned wood. The extent of seasoning 
can be determined by separating the oil from the crystals with 
methyl alcohol in which the latter are sparingly soluble. 

Fat, Resin, and Wax. — It is a popular belief in the South that 
segregation of resin takes place in the stumps and fallen trunks of 
pines. This is erroneous. In actuality, the less resinous por- 
tions soon disappear through decay, leaving only the richer wood 
which is resistant. Nordenskjold 4 found 15 to 22 per cent of 
resin in pine stumps that had stood for 10 years. The average 
resin content of 15 trees was 4.2 per cent for the sap wood and 15 
per cent for the heart wood. 

The resin content of some species used for paper making is 
given below: 5 



Ether 


Alcohol 


extract, 


extract, 


per cent 


per cent 


0.40 


0.36 


0.24 


0.36 


0.85 




1.20 


1.35 


0.87 


1.71 


2.70 


0.99 


3.24 


1.71 


2.37 


0.77 



Total 
extract, 
per cent 



1. White spruce, seasoned 

2. Black spruce, seasoned 

3. Balsam, seasoned 

4. Balsam, green 

5. Balsam, decayed 

6. Jack pine, seasoned, large tree 

7. Jack pine, green, small tree . . . 

8. Jack pine, rafted, seasoned 



0.76 
0.60 

2.55 
2.58 
3.69 
4.95 
3.14 



1 R. Kanehira, J. Forestry, 19 (1921), 736-739. 

2 R. Romanis, J. Chem. Soc, 51 (1887), 868-871. 

3 A. C. Sircar, J. Soc. Chem. Ind., 35 (1916), 452-454. 

4 1. Nordenskjold, Arkiv Kemi, Mineral. Geol., 4, 28 (1912), 1-21. 
5 F. Barnes, Chem. Met. Eng., 28 (1923), 504. 



66 



CHEMISTRY OF CELLULOSE AND WOOD 



The following extractions were made on branches 2 to 4 years 
old. 1 



Ether, Alcohol, 
per cent per cent 



Water. 
per cent 



Total, 
per cent 



Water 

only, 

per cent 



Pinus silvesiris, wood 
Pinus silvestris, bark. 
Pinus abies, wood 
Pinus abies, bark. . . . 



7.87 


4.19 


3.55 


15.61 


19.58 


14.76 


13.27 


47.61 


4.40 


2.12 


2.45 


8.97 


10.66 


14.22 


13.43 


38.31 



6.62 
28.50 

4.96 
21.95 



Bergstrom 2 found that the soap obtained in cooking pine wood 
by the sulphate process consisted of a mixture of resin and fatty 
acids. Pine (P. silvestris) contained more fat than spruce 
{Picea excelsa). Fats are present only in winter, being converted 
into starch in spring. The heartwood is richest in resin and the 
sap wood in fats. The sapwood of pine felled in winter contained 
0.3 to 1.5 per cent of fats. The fats consisted of the glycerides of 
oleic, linolenic, and palmitic acids, and contained a phytosterol 
(0.05 per cent), m.p. 133.5°, sparingly soluble in alcohol. 
Glycerine was detected in the ether-alcohol extract of the 
sapwood. 

The following amounts of fat and resin are reported by 
Schwalbe: 3 



Fat, per cent 



Resin, per cent 



Spruce, seasoned 
Pine, seasoned. . . 



0.50 
1.43 



0.48 
1.11 



The fat consisted principally of the glyceride of oleic acid, 
with small amounts of the esters of linolic and linolenic acids. 4 
The ether extract and alcohol extract of sulphite pulp contained 
63 and 74 per cent respectively of fat. It is this fat which 

1 O. Aschax and O. Raxtalaixex, Brennstoff-Chemie, 4 (1923), 101. 

2 H. Bergstrom, Papier-Fabr. Fest u. Auslandsheft, 9 (1911), 76-80; 
Pa-yier-Fabr., 11 (1913), 730. 

3 C. G. Schwalbe, Forst u. Jagdwesen, 47 (1915), 99. 

4 C. G. Schwalbe, I.e.; R. Sieber, Z. angew. Chem., 29 (1916), 429. 



THE COMPOSITION OF WOOD 



67 



causes " resin spots" in paper 1 and " pitch trouble" on the paper 
machine. 

The resins and fats change with age. Richter 2 found that as 
the wood seasoned, the ether extract decreased and the alcohol 
extract increased, though the latter was insufficient to balance 
the loss of ether extract. The acid and saponification numbers of 
the extractives increased with the aging of the wood, while the 
ester number decreased. 3 Oxidation promoted these changes. 

The amount of resin and fat extractable from wood decreased 
from one-half to two-thirds through aging for 4 to 18 months. 4 
The ether extract from pine wood was 6.3 per cent 6 days after 
comminution of the wood, and 2.1 per cent 127 days afterwards. 
An experiment with chips showed a reduction of the ether extract 
from 5.9 to 2.6 per cent in 15 days. The decrease in solubility is 
attributed to oxidation or condensation, or both. Without doubt 
the decrease is due to the formation of linoxyn, which is insoluble 
in most organic solvents and which retards solution of the resin. 

Sawdust from Rafted Pine Wood 



Days 
aged 


Solvent 


Extract, 
per cent 


Resin, 
per cent 


Fat, 
per cent 


85 
127 


Ether 

Ether 


6.9 
2.5 
7.3 
2.5 


67.2 
87.9 
75.8 
92.5 


32.8 
12.1 


85 
127 


Benzene 

Benzene 


24.2 

7 5 









The amount of fat and resin in fresh and seasoned coniferous 
woods and in wood pulps has been extensively studied by Sieber. 5 

Walberg 6 found that drying the wood at 105° materially 
decreased the benzene extract but had no effect on the alcohol 
extract. Storage in air or nitrogen made no difference. Storage 
for 80 days showed no difference in resin content over a storage 

1 C. G. Schwalbe, Chem. Ztg., 38 (1914), 926. 

2 E. Richter, Wochbl. Papierfabr., 44 (1913), 2486. 

3 C. G. Schwalbe and H. Grimm, Wochbl. Papierfabr., 44 (1913), 3247. 
4 C. G. Schwalbe and W. Schulz, Z. angew Chem., 31, 1 (1918), 125-127; 

Chem. Zeit., 42 (1918), 229-230. 

5 R. Sieber, "Harz der Nadelholzer," Schriften des Vereins der Zellstoff-u. 
Papierchemiker, Berlin (1914), 109 pp. 

6 H. E. Wahlberg, Papier-Fabr., 20 (1922), 1136. 



68 CHEMISTRY OF CELLULOSE AND WOOD 

period of 10 days. The resin was very irregularly distributed in 
the wood and the statement of Mayr 1 that it occurs in largest 
amounts on the south side of the tree could not be confirmed. 

Cutin. — The occurrence of cutin in wood is questionable. 
Konig and Rump 2 have reported 0.14 and 0.16 per cent of cutin 
in the crude fiber of beech and fir wood, respectively. The 
wood was freed from cellulose by treatment with 72 per cent 
sulphuric acid; the residue consisted of lignin and cutin from 
which the lignin was removed by oxidation with hydrogen 
peroxide and ammonia. Cutin is a wax containing 69 to 70 
per cent carbon and 9 to 12 per cent hydrogen; saponification 
gives compounds agreeing with nonylic or capric acid, and cetyl 
or octadecyl alcohol. Of several woods examined by Dore, 3 
oak alone gave a substance that might have been cutin. 

Irritant Constituents. — Several species of wood contain 
compounds such as alkaloids, glucosides, and resins which irritate 
the skin and mucous membrane, and produce heart depression 
and paralysis of the motor centers. Workmen vary greatly in 
susceptibility. For this reason and the difficulty in isolating 
the active constituent or constituents in a pure state, there are 
contradictory reports on the physiological effects. 

The active principle in East Indian satinwood (Chloroxylon 
swietenia) is the alkaloid chloroxylonine, C 22 H230 3 N; m.p. 182 to 
183°. 4 Nestler 5 was unable to obtain irritation with either the 
wood or its extracts. 

According to Nestler, liquid amber or red gum (Liquidambar 
styraciflua L.) from North America is not directly irritant, but 
the ether extract is strongly so. The alcohol, water, and chloro- 
form extracts are inactive. The writer extracted an authentic 
sample of wood of this species with ether and obtained a solution 
with a green fluorescence. Application to the forearm of the 
brown, wax-like mass obtained on evaporation of the solvent did 
not produce a trace of inflammation after contact for 9 hours. 

X H. Mayr, "Das Harz der Nadelholzer," Berlin (1894). 

2 J. Konig and E. Rump, "Chemie und Struktur der Pnanzen-Zellmem- 
bran," (1914), p. 52; Z. Unters. Nakr. Genussm., 28 (1914), 177-222; cf. 
W. Suthoff, Ibid., 17 (1909), 662-663. 

3 W. H. Dore, /. Ind. Eng. Chew,., 11 (1919), 556. 

4 S. J. M. Auld, J. Chem. Soc, 95 (1909), 964-968. 

5 A. Nestler, Ber. botan. Ges., 29 (1911), 672. 



THE COMPOSITION OF WOOD 



69 



The irritant action of various species of cocobolo (Dalbergia) 
is apparently due to an oil. 1 Nestler 2 states that the active 
substance is soluble in water, but more easily in alcohol and 
benzene. 

The species given in the table below have been examined by 
Matthes and Schreiber. 3 Lapachol and lapachonone are not 
irritant. Wood containing lapachol shows small cherry-red spots 
when moistened with 0.1N alcoholic potash. Lapachonone gives 



Species 



Active 
principle 



Compounds identified 



Surinam greenheart (Bignonia 

lucoxylon) 

Guiana greenheart (Nectandra 

rodiaei) 

Moah wood, East Indies (Illipe 

longijolia) 

Teak (Tectonia grandis) 

Native teak, Australia (Flindersia 

australis) . 
Lapacho wood, South America 

(BignoniacecBj 

Tecoma araliacea 




Resin 

Resin 

(Not toxic) 



Resin 
Resin 



Lapachol, 1 C29H26O4 

(Lapachol absent) 

Lapachol; lapachonone 

Flindersin, C23H26O7N2, an 
alkaloid; resin; oil 

Lapachol 
Lapachol 



1 C. A. Oesterle, Arch. Pharm., 254 (1916), 346. 

an intense indigo-blue solution with sulphuric acid. According 
to Baker, 4 wasahba wood, which is used for fishing rods, contains 
a resinous substance giving a red color with alkalis. It is about 
as sensitive as phenolphthalein. The resin may contain lapachol. 
Other species reputed to possess irritant properties are: 
boxwood (Buxus sempervirens and B. macowani); African box- 
wood (Gonioma kamassi); satinwood (Zanthoxylon flava); green 
ebony (Brya ebenus); West Indian sabicu wood (Lysiloma sabicu); 
ebony, Borneo rosewood; magneta rosewood; olive wood; and 
mahogany. 5 

1 G. H. Garratt, J. Forestry, 20 (1922), 479. 

2 A. Nestler, Ber. botan. Ges., 29 (1911), 672. 

3 H. Matthes and E. Schreiber, Ber. botan. Ges., 24 (1914), 385-444; 
contains an extensive medical bibliography. 

4 G. B. Baker, Chem. Neivs, 116 (1917), 139. 
5 G. H. Garratt, J. Forestry, 20 (1922) 479-487. 



CHAPTER III 
LIGNIN 

Lignin represents the non-carbohydrate portion of lignified 
tissue after it has been freed from tannins, resins, fats, and similar 
secondary constituents. The study of lignin is rendered exceed- 
ingly difficult owing to the impossibility of isolating it unchanged. 
Furthermore, it is not uniform. There is a great difference in the 
methoxyl and acetyl contents of lignin from various species of 
woods, particularly between hardwoods and softwoods. Some 
variation exists even in lignin from different portions of the same 
tree. Lignin is easily oxidized by nitric acid. Halogens pro- 
duce oxidation and substitution, the halogenated product being 
readily soluble in alkalis. It is soluble in acetyl bromide in the 
cold and in alkalis, sulphurous acid, and sulphite solutions at high 
temperatures. Treatment of lignified tissue with strong mineral 
acids, which dissolve the carbohydrates, leaves a residue of 
lignin. 

The structure of lignin is still uncertain. In many of its 
reactions lignin closely resembles the tannins, though no convinc- 
ing evidence of a cyclic structure has been produced. Pyrolytic 
methods are useless as a means of determining constitution. 
Nitration, chlorination, sulphonation, oxidation, and reduction 
give either indefinite products, or substances furnishing only 
questionable information on structure. Furan, benzofuran, 
pyrone, benzenoid, and aliphatic structures have been proposed. 
It has been definitely established that lignin contains methoxyl, 
acetyl, and hydroxyl groups, but the presence of ketonic and 
aldehydic groups has not been satisfactorily proved. 

The hydroxyl groups, combined for the most part as esters 
and ethers, appear to be both phenolic and alcoholic. Assuming 
for the lignin of wood a nucleus of the order of C45, it contains 
eight to ten hydroxyl groups, of which three to five are combined 
as methyl ethers, while the remainder are mostly acetylated. 

70 



LIGNIN 71 

One aldehydic and two ketonic groups may be present. The 
presence of formyl and carboxyl groups is doubtful. 

The name lignin, in the sense in which we at present use it, 
appears to have been first employed by F. Schulze 1 in 1857. He, 
though not the first to do so, distinguished cellulose from lignin 
or incrusting matter by its resistance to oxidizing agents. By 
indirect analysis, lignin was assigned the formula C38H24O20 and 
considered to be a pure material. 

Dragendorff 2 in his classic work states that the vasculose of 
Fremy agrees in the main with his lignin. He mentions that 
lignin cannot be separated from cellulose without decomposition, 
so that it is impossible to obtain direct proof whether it consists 
of one or several chemical individuals. On the basis of the work 
of his students, he is inclined to the opinion that cellulose is 
associated with a single definite substance, lignin. Several 
species of wood were examined by Stackmann; 3 after exhaustion 
with neutral solvents, dilute alkali and acid, the composition of 
the lignin was determined by ultimate analysis of the purified 
material before and after treatment with chlorine. The lignin of 
dicotyledons contained : 

Per Cent 

C. 53.1-59.6 

H 4.4- 6.3 

O 34.1-38.9 

The results are subject to correction for "wood gum" that had 
not been discovered. 

Schuppe 4 found from 3.25 to 7.09 per cent of wood gum in 
various angiosperms, and by making the necessary deductions 
obtained for most woods about 17 per cent of lignin. Wood was 

1 F. Schulze, Chem. Centr., 28 (1857), 321-325. Czapek ("Biochemie 
d. Pflanzen," vol. I (1913), p. 682), without giving a reference, credits de 
Candolle with priority in the use of lignin. A. de Candolle in his " Organo- 
graphie Vegetale," Paris (1827), and "Introduction a l'Etude de la Botan- 
ique," Paris (1835), uses the terms couches ligneuses, portion ligneuse, and 
corps ligneux in the sense of Malpighi and other earlier writers as represent- 
ing the woody portion of the tree. 

2 G. Dragendorff, " Plant Analysis," London (1884). 

3 Stackmann, "Studien iiber die Zusammensetzung des Holzes," 
Diss. Dorpat (1878). 

4 N. Schuppe, "Beitrage zur Chemie des Holzgewebes," Diss. Dorpat 
(1882). 



72 CHEMISTRY OF CELLULOSE AND WOOD 

treated with water, alcohol, and dilute sodium hydroxide, and 
analyzed. He proposed for lignin the formula Ci 9 Hi 8 8 , and 
for wood 5C 6 Hi O5 + Ci 9 Hi 8 8 . 

Lignin, Wood, 

Pek Cent Per Cent 

C 60.56 45.4 

H 4.66 5.9 

O 34.80 48.7 

In 1889, Cross and Bevan 1 advanced an empirical formula for 
jute lignocellulose and formulated its reaction with chlorine. The 
equation does not balance, but the chlorine derivative of lignin 
was assigned the formula C19H17CI4O9. Later 2 the compound 
was given as CigHisCUOg, containing uniformly 26.7 per cent of 
chlorine, the parent substance being C19H22O9. 

Lignin appears to be profoundly modified by all methods 
employed for its isolation; hence it seems best to group the 
literature for the most part under the process whereby the lignin 
was obtained. 

Lignin Isolated with Sulphite Solutions. — The industrial 
importance of the sulphite process for pulping wood has led to 
an extensive investigation of the constitution of lignin through a 
study of the reaction whereby the lignin is dissolved. Lindsey 
and Tollens 3 used a waste sulphite liquor obtained by cooking 
coniferous wood by the Ritter-Kellner process. Lead acetate 
gave a precipitate which on decomposition with sulphuric acid 
and precipitation with alcohol had the empirical composition 
C26H30SO12. The filtrate on evaporation gave a gummy com- 
pound C26H30SO12. When the original liquor was treated with 
bromine, the derivative, C 2 6H 2 8Br 4 SOii, was thrown down, while 
with hydrochloric acid the precipitate was represented by C 2 6H 30 - 
SO10. Distillation with hydriodic acid showed the presence of 
two methoxyl groups. The sulphur was present as — S0 3 H. It 
was concluded that the ligneous body had the composition 
C24H 2 4(OCH 3 )20io or C 2 4H26(OCH 3 )20io. Fractional precipita- 
tion of sulphite liquor with alcohol by Seidel 4 failed to produce 
constitutional differences in the precipitates. A carefully puri- 

1 C. F. Cross and E. J. Bevan, /. Chem. Soc, 55 (1889), 207. 

2 "Researches" (1895-1900), pp. 126, 134. 

3 J. B. Lindsey and B. Tollens, Ann., 267 (1892), 341. 

4 H. Seidel, Mitt. k.k. techn. Gewerbe-Mus., 7 (1897), 219. 



LIGNIN 73 

fied fraction had the composition C, 56.27 per cent; H, 5.87 per 
cent; and S, 5.52 per cent. This analysis corresponds closely 
with the formula C26H30SO12 of Lindsey and Tollens. 

When sulphite liquor was treated with bleaching powder in 
the presence of hydrochloric acid, a chlorine derivative was 
obtained, which, after purification with alcohol and ether, had 
the composition C26H29CISO12. 1 Streeb 42 obtained the formula 
CisH^SOio for ligninsulphonic acid from sulphite liquor. 

To isolate the lignin from sulphite liquor, Klason 3 heated the 
latter with calcium chloride whereby about half of the calcium 
salts of lignin was precipitated. Purification was effected by 
treating the salts with sulphuric acid, and then with alcohol to 
throw down the calcium sulphate. After evaporation of the 
alcohol from the nitrate, the free sulphonic acids were taken up in 
water, neutralized with barium hydroxide, and the barium salt 
precipitated with alcohol. Analysis of the salt showed the compo- 
sition C4oH 44 Oi 7 S2Ba. The molecular weight was about 6,000, 
so that lignin would be expressed by the formula (C^H^OnV 
The purified salt gave none of the lignin color reactions. It 
contained 11.5 per cent methoxyl, corresponding to 4CH 3 for 
each C 4 o. Treatment of the calcium salt with p-bromobenzoyl- 
chloride pointed to the presence of one hydroxyl group. It did 
not condense with phenylhydrazine or benzylphenylhydrazine ; 
however, some evidence of the presence of an active carbonyl 
group was drawn from heating the ligninsulphonic salt with 
calcium bisulphite solution whereby a loose combination with 
two molecules of S0 2 was obtained. The presence of an ethylene 
linkage was shown by the addition of a molecule of iodine. 

Klason advanced the hypothesis that lignin consists of a 
condensation product of coniferyl and oxyconiferyl alcohol. 
Coniferyl alcohol is resinified readily by mineral acids, as is 
lignin, when a sulphite cook is deficient in lime. Lignin combines 
with four molecules of S0 2 , two of which are present in the form of 
the sulphonic acid. The third is more loosely combined and, 
although not titratable with iodine, gradually splits off as sul- 

1 H. Krause, Chem. Ind., 29 (1906), 217. 

2 E. Streeb, "Ueber Derivate des Lignin's," Diss. Gottingen (1892), 
31 pp. 

3 P. Klason, " Zusammensetzung des Fichtenholzes " (1910), p. 15. 



74 CHEMISTRY OF CELLULOSE AND WOOD 

phuric acid on standing in the air. The fourth molecule is the 
most weakly combined and can be titrated with iodine. Since 
the salt C^H^OiyS^a can combine with only two atoms of 
iodine, an active carbonyl group must be present as a glycide, 
— CH 2 .CH.CH 2 , as it does not react with phenylhydrazine but 

V 

does with sulphur dioxide. A — CH 2 .CO.CH 3 group is not present, 
since no acetic acid is formed by oxidation with nitric acid. 
It is assumed that in each C 4 o group there are present four 
CH3O and three OH groups. Experimental evidence indicated 
only one or two hydroxyl groups, but it is probable that a 
— C.COH group passes to — CH.CO — . Since the lignin isolated 
directly from spruce contained 7.01 per cent methyl corre- 
sponding with two CH3O groups and the salt C 40 H 44 Oi 7 S 2 Ba 
contained four CH 3 groups, there must be at least two forms of 
lignin in sulphite liquor, that not precipitated by calcium chloride 
being practically free from methoxyl. There is considerable 
analogy between lignin and the tannins; in fact, lignin may be 
considered as an insoluble tannin, the side groups being in the 
1, 3, 4, and 5 position, as in gallic acid, thereby explaining its 
ready oxidizability. 

In subsequent papers, Klason's hypothesis of the composition 
and constitution cf lignin has undergone many modifications. 
In 1916, lignin was formed as follows: 1 

lC 10 H 12 O 3 + 3CioHi 2 4 + 6C 9 Hi O 4 = C 94 H 88 29 + 10H 2 O. 

Coniferyl Oxyconiferyl Trioxycinnamyl Lignin 

alcohol alcohol alcohol 

The lignin precipitated from sulphite liquor with calcium 
chloride in the form of calcium ligninsulphonate has the formula 
C40H40O11, and that remaining in solution, C 54 H 48 0i 8 . 

The calcium salt of ligninsulphonic acid precipitated by calcium 
chloride is formed as follows: 2 
lC 10 H 12 O 3 + 3C 10 H 12 O 4 + lCa(S0 3 H) 2 = C 40 H 41 O 18 S 2 Ca + 3H 2 0. 

Coniferyl Oxyconiferyl 

alcohol alcohol 

This corresponds with the formula C 40 H 42 Oi 2 with four CH 3 
groups. The calcium salt reacts with napthylamine to give a 

1 P. Klason, Svensk Papers-Tid., 17 (1916); Arkiv Kcmi, Mineral. Geol., 
6, 15 (1917), 6. 

2 P. Klason, Arkiv Kemi, Mineral. Geol, 6, 15 (1917), 1-21. 



LIGNIN 75 

yellow precipitate. With the salt not precipitated by calcium 
chloride there is obtained the compound C59H66O17S2N3. It may 
be formed as follows : 
lC 10 H 12 O 3 + lC 10 H 12 O 4 + 1C 9 H 8 4 + 2H 2 S0 3 + 3C 10 H 9 N = 

Coniferyl Oxyconiferyl Caffeic Sulphurous Napthylamine 

alcohol alcohol acid acid 

C 5 9H 57 1 4S 2 N3 1 + 3H 2 0. 
This compound corresponds with the formula C 29 H 26 8 , so that 
the total lignin has the formula C6 9 H 6 80 2 o with six CH 3 groups. 
Klason believes that the lignin of fir is formed by condensation of 
oxy and methylated derivatives of cinnamic alcohol and related 
aldehydes and acids. 

On reviewing the paper, the impression is gained that 
compounds were chosen in such a way that they could be con- 
densed to fit the empirical formula obtained on analysis, rather 
than on the basis of experimental evidence. The fact that wood 
on oxidation with ozone 2 does not give vanillin or one of its 
derivatives is a strong indication that lignin does not contain a 
coniferyl alcohol group. 

The jS-napthylamine ligninsulphonate obtained from the calcium 
salt precipitated with calcium chloride exists in two forms, thus 
supporting, according to Klason, 3 the theory that lignin has a con- 

,CH:CH.CH 2 .OH (1) 
stitution similar to coniferyl alcohol, C 6 H3^-OCH 3 (3), 

X OH (4) 

which is a substituted allyl alcohol. A part of the allyl alco- 
hol complex exists in the form of the aldehyde and this reacts 
with sulphurous acid and napthylamine as follows: 

1. R.CH:CH.CHO + H.S0 2 .OH = R.CH 2 .CH(S0 2 .OH).CHO 

(A). 

2. A + C 10 H 7 NH 2 = R.CH 2 .CH(SO 2 .O.NH 3 .Ci H 7 )CHO (B). 

S0 2 .0 

3. B->R.CH 2 .CH<( I (C). 

X CH:NH.C 10 H 7 

1 Owing to the use of an incorrect formula (C 10 H ]0 N) for napthylamine 
in the original (p. 7), the compound is stated to be C59H60O14S2N3 and the 
combustion figures compared with this formula. 

2 C. Doree and M. Cunningham, J. Chem. Soc, 103 (1913), 677. 

3 P. Klason, Ber., 63 (1920), 706-711; 1862-1863; 55 (1922), 448. 



76 



CHEMISTRY OF CELLULOSE AND WOOD 



The white precipitate B first obtained rearranges in acid 
solution to give the intense yellow cyclic salt C. The positions 
1, 3, and 4 are based on the rather slender evidence furnished by 
the presence of vanillin in sulphite liquor and the formation of 
w-propyl creosol by the destructive distillation of wood. The 
napthylamine salts prepared from the sulphite liquor in which 
various woods were cooked had fundamentally the same appear- 
ance and properties as that from fir. Their composition is shown 
below: 





Spruce, 
per cent 


Pine, 
per cent 


Birch, 
per cent 


Oak, 
per cent 


Aspen, 
per cent 


Willow, 
per cent 


Beech, 
per cent 


N 

S 


2.07 
5.22 


1.95 
5.34 


2.40 

4.84 


1.82 
5.02 


1.87 
4.84 


2.04 
5.43 


2.00 
5.26 







Hintikka 1 considers untenable the belief that there occurs in 
lignin an acrolein group which combines with bisulphites to form a 
sulphonic acid giving the yellow cyclic salt C. The difference in 
behavior of the salts B and C on treatment with dilute caustic 
soda as reported by Klason could not be substantiated. It is 
apparently a simple salt. 2 

The lignin of spruce wood exists in two forms, 3 63 per cent 
being acrolein- or a-lignin, and 37 per cent carboxyl- or /3-lignin. 
<*-Lignin contains an acrolein group in the form of coniferyl 
aldehyde. Its sulphonic acid forms cyclic aryl ammonium salts 
and its calcium salt is precipitated by calcium chloride. The 
calcium salt has the composition: 







Calculated 


Found 


C40 


480 
44 

288 
64 
40 

916 
124 


52.41 
4.79 

31.45 
6.98 
4.37 


52.82 
5.08 


Ols 




So 


6.58 


Ca 


4.15 






4CH 3 


100.00 
13.53 


12.19 







1 S. V. Hintikka, Cellulosechemie, 2 (1921), 63; 4 (1923), 93. 

2 C. T>on±ii and L. Hall, J. Soc. Chem. Ind., 43 (1924), 261T. 

3 P. Klason, Ber., 53 (1920), 1864-1873. 



LIGNIN 77 

Klason believes that the above formula should be cut in half. 
Furthermore, it is subject to a correction, for the acetic acid split 
off during heating with sulphite liquor. When 45 grams of wood 
were cooked with sulphite liquor until all the lignin had been 
removed, 0.945 gram of acetic acid was formed, while subsequent 
hydrolysis with sulphuric acid yielded an additional 0.21 gram. 
Of the total 1.155 grams of acid, 10 per cent consisted of formic 
acid. Spruce wood digested with N H 2 S0 4 on the water bath 
for 24 hours gave 2.1 per cent acetic acid. Assuming the presence 
of one acetyl group, and that a-lignin constitutes about 63 per 
cent of the total lignin (30 per cent), the theoretical yield would 
be 2.86 per cent of acetic acid. The corrected formula for 
a-lignin would be C22H22O7, having probably the constitution 
of a flavone. The hydroxyl groups in a-lignin are either 
methylated or acetylated. 

jS-Lignin contains an acrylic acid group R.CH:CH.COOH 
and has the formula Ci 9 Hi 8 9 . According to calculation, the 
wood should give 3.78 per cent CH 3 0, when actually 4.1 per cent 
was found. This is attributed to the formation of some ethyl 
iodide from the complex R.CH:CH.CH 2 OH. Coniferin gave 
9.73 per cent CH 3 against the calculated 8.20 per cent. It has 
been shown, however, that methoxyl is the sole alkoxyl group in 
wood and lignin. 1 

Further analyses of the /3-naphthylamine and calcium salts 
of /?-ligninsulphonic acid indicate the formula CioHigOr instead 
of CigHigOg. /3-Lignin may be formed by condensation of one 
molecule of coniferyl aldehyde with one molecule of caffeic acid. 2 

Spruce wood after extraction with ether to remove resins 
was extracted alternately with water and alcohol, which removed 
12 per cent of the weight of the wood. 3 The residue from evap- 
oration of the alcohol extract was extracted with chloroform; 
the soluble portion contained 66.14 per cent C, and 6.49 per 
cent H; the insoluble portion 59.97 per cent C, and 5.66 per cent 
H. The first substance on oxidation gave a vanillin odor and 
was considered from analyses and molecular-weight determina- 
tions to be in good agreement with coniferyl alcohol, CioHi 2 3 ; 

1 E. Hagglund and B. Sundroos, Biochem. Z., 146 (1924), 221. 

2 P. Klason, Ber., 56 (1923), 300. 

3 P. Klason, " Zusammensetzung des Fichtenholzes," p. 30. 



78 



CHEMISTRY OF CELLULOSE AND WOOD 



while the second was a dimeric form, (CioHi 2 4 ) 2 , of oxyconi- 
feryl alcohol. The portion insoluble in chloroform when com- 
pletely freed from water by heating at 130° contained 63.94 
per cent C and 5.74 per cent H. 1 Since about 65 per cent com- 
bines with jS-napthylamine the product giving the typical reac- 
tions for lignin contains two-thirds a-lignin and one-third 
/3-lignin. The lignin thus extracted directly from wood is sub- 
stantially in agreement with the lignin obtained from sulphite 
liquor. Lignin and its sulphonic acid appear to be optically 
inactive. 

The homogeneity of calcium a-ligninsulphonate was shown 
by the uniformity of the analyses of the /3-napthylamine salts 
obtained by : treating waste sulphite liquor with calcium chloride ; 
fractional precipitation of the waste liquor with sodium chloride ; 
and from the portion remaining dissolved. 2 When a-lignin- 
sulphonic acid is oxidized with hydrogen peroxide, a-napthyla- 
mine precipitates a salt, C40H38O11SN2. 

C20H20O6 + H2SO3 + 2C10H7.NH2 + O2 — H 2 = C40H38O11SN2. 





C 


H 


S 


N 


Calculated 

Found 


63.64 
63.29 


5.08 
5.33 


4.25 
4.20 


3.71 
3.49 







It is concluded from the composition of the salt that an alde- 
hydic group has been oxidized to a carboxyl group and a meth- 
ylene group to a ketonic group. The presence of two hydroxyl 
groups is shown by the preparation of a diacetyl derivative of 
jS-napthylamine-a-ligninsulphonate; however, since only one 
group can be methylated, one is probably attached to a benzenoid 
nucleus, the other to an aliphatic group. 

The flavone structure proposed for a-lignin has been abandoned 
in favor of the following formula, 3 which contains a benzofuran 
ring. This structure explains the formation, on destructive distil- 
lation, of guaiacol, catechol, m- and p-cresols, methyl-, ethyl-, and 

1 P. Klason, Ber., 55 (1922), 455-456. 

2 P. Klason, Ber., 55 (1922), 448-455; Svensk Kern. Tid., 34 (1922), 4-17. 

3 P. Klason, Ber., 56 (1923), 300. 



LIGNIN 79 

n-propylcreosols, methyl alcohol, allyl alcohol, and 2-methyl- 
furan. It has been shown that a-lignin is a chemical compound 
in which can be quantitatively determined an acrolein group, 
two methoxyl groups, a phenolic and alcoholic hydroxyl group, 
an aliphatic methylene group, and oxidic oxygen. A body of 
the above constitution must show two aromatic nuclei connected 
by a cumarone or furan ring. Klason also points out the close 
relationship of the above formula to gambier-catechu, 2.4.6.3'.4' 
-p3ntamethoxy-a, 7-diphenylpropane, synthesized by Freuden- 
berg, 1 in view of the opinion held by many investigators of the 
similarity of lignin to the tannins. 

CH3O.C O CH 

HC C CH CH 2 — C C.OCH3 

I II I II I 

OHC.CH:CH.C C CH.OH HC C.OH 

\/ \/ 

CH CH 

An apparently close relationship exists between certain alcohol- 
soluble resins, tannins, and lignin. 2 Spruce needles gave with 
95 per cent methyl alcohol an extract separable into three main 
fractions, a crude fat soluble in ether, a crude resin insoluble in 
ether but soluble in alcohol, and a syrup soluble in water. The 
three fractions contained bodies derived from hydroxycinnamic 
acids, aldehydes, and alcohols, that differed from each other in 
various degrees of methylation, hydroxylation, hydrogenation, 
and condensation. The crude fat contained 29 per cent of 
abiephyllic acid having the properties of a tannin. The crude 
resin contained bodies having aldehydic and ketonic properties 
and behaving like a tannin in that an alcoholic solution precipi- 
tates gelatin, and an alkaline solution on acidification precipitates 
a "red" similar to the phlobaphenes. 

The crude resin resembles the a-lignin and the water-soluble 
syrup the /3-lignin of Klason. A closely related series of tannic 
acids isolated from the syrup represents ketoacids derived from 
hydroxycinnamic acid and related to p-cumarylferulic acid or 

1 K. Freudenberg, Ber., 53 (1920), 1416. 

2 A. C. von Euler, Cellulosechemie, 2 (1921), 128-135; 3 (1922), 1-7. 



80 CHEMISTRY OF CELLULOSE AND WOOD 

feruylferulic acid, and to caffeic acid. Coniferous wood, after 
a preliminary extraction with benzene, yields to alcohol a resin 
considered to be a soluble form of lignin. 1 In view of the 
presence of various cinnamic acid derivatives in the resin of 
Pinus laricio, 2 it may be questioned whether there is any genetic 
relationship between these substances and lignin. 

When sulphite liquor is heated at 60 to 70° with 21 per cent 
of its weight of sodium chloride, /3-lignin-S-acid remains in 
solution while a-lignin-S-acid is precipitated. 3 The latter 
differs materially from Klason's ligninsulphonic acid obtained 
by precipitation with CaCl 2 . The a-lignin-S-acid is a mixture 
of ligninsulphonic acids of high molecular weight (about 900); 
since conductivity measurements showed one ionizable hydrogen 
atom for each atom of sulphur, a carboxyl group is not present. 
Most of the sulphur dioxide is loosely combined. The product 
purified by repeated precipitation with sodium chloride showed 
no constant ratio of carbon to sulphur. Products of constant 
composition were not obtained with aromatic amines; as one 
nitrogen was present for each sulphur atom, the acidity is entirely 
due to the sulphur group. The composition of the sulphur-free 
compound varied from C28H34O8 to C34H44O11. Since acetic 
acid is split off by alkalis, a-lignin may consist of a mixture of 
substances differing from a parent substance in the wood through 
the loss of formyl and acetyl groups. 

Holmberg 4 on extracting a liter of sulphite liquor with ether 
obtained 0.2 gram of a lactone, C 2 oH 2 o0 6 , called sulphite liquor 
lactone. It crystallizes in plates or prisms and melts at 250 to 
255°. In alcoholic solution it gives with ferric chloride a green 
coloration changing to brown; with phloroglucinol and hydro- 
chloric acid there is no reaction, but an alkaline solution of the 
lactone after exposure to air gives an intense cherry-red colora- 
tion with this reagent. Ligninsulphonic acids from sulphite 

1 A. C. von Euler, Cellulosechemie, 4 (1923), 4. 

2 M. Bamberger, Monatsh., 12 (1891), 445; 15 (1894), 505. 

3 K. H. A. Melander, Cellulosechemie, 2 (1921), 41-43, 69-73; J. Soc. 
Chem. Ind., 38 (1919), 625A. 

4 B. Holmberg, Svensk Kern. Tid., 32 (1920), 56-67; Ber., 54 (1921), 
2389-2406. 



LIGNIN 



81 



liquor exhibit a similar behavior. The lactone is presumably 
that of diguaiacyltetramethylene-carbinol carboxylic acid: 

C 6 H 3 (OCH3)(OH).CH— CH.CH 2 OH 

C 6 H 3 (OCH 3 )(OH).CH— CH.COOH 

Sulphonation of the lactone and subsequent treatment with water 
yields a monosulphonic acid, m.p. 172 to 173°. Dimethyl- 
sulphate gives a-dimethyl-sulphite-liquor-lactone, d 8 Hi 2 2 .- 
(OCH 3 ) 4 , in the form of silky needles, m.p. 179 to 180°; [a] l £° = 
— 100.9°. The a-compound on treatment with sodium ethoxide 
gives an inactive /3-dimethyllactone. 1 

The lactone is found in greater quantity in the first part of 
the period of the digestion of spruce wood with sulphite liquor 
than towards the end. 2 It was not obtained from spruce bark 
or pine wood. 

No success has yet been attained in isolating definite lignin- 
sulphonic acids. Honig and Spitzer, 3 by the fractional precipi- 
tation with alcohol of the calcium or barium salts from evaporated 
sulphite liquor, obtained fractions of variable composition 
roughly divisible into three classes: (1) C4 3 H 5 oOi 8 S 2 Ba, containing 
four CH 3 groups; (2) CioH^OisS^a, containing one CH 3 
group; and (3) C74Hii 4 48 S 2 Ba, containing three CH 3 groups, 
and representing the portion remaining unprecipitated by 
alcohol. The salts on ultimate analysis showed the following 
composition : 





(1) 
per cent 


(2) 
per cent 


(3) 
per cent 


c 


49.39 

4.75 

6.39 

10.81 

11.15 


47.41 
4.96 
6.46 

11.29 
2.95 


44 43 


H 


5 75 


S 

Ba 

CH 3 


3.65 
6.90 

4 86 







It is conceivable that salt 2 is formed from salt 1 by elimination 
of three methoxyl groups. The free lignin acids were soluble 
in water, glacial acetic acid, and alcohol, and liberated carbon 

1 B. Holmberg and M. Sjoberg, Ber., 54 (1921), 2406-2407. 

2 S. V. Hintikka, Cellulosechemie, 4 (1923), 93. 

3 M. Honig and J. Spitzer, Monatsh., 39 (1918), 1-14. 



82 CHEMISTRY OF CELLULOSE AND WOOD 

dioxide from carbonates and acetic acid from acetates. In 
their opinion, though the ligninsulphonic acids so far isolated 
are not uniform substances but mixtures, the several sulphonic 
acids are best characterized by their methoxyl content. 

McKee and Barsky 1 fractionally precipitated the calcium 
salts of ligninsulphonic acids from sulphite liquor, later convert- 
ing them into the barium salts by Klason's method, and obtained 
fractions of very variable composition. The carbon content 
varied from 59.7 to 55.1 per cent and the hydrogen from 6.4 
to 5.9 per cent. 

Konig 2 isolated a barium ligninsulphonate of the composition 
C46H 4 o027S 2 Ba by spraying concentrated sulphite liquor into 50 
per cent sulphuric acid, neutralizing the precipitated lignin- 
sulphonic acid with barium carbonate, and precipitating the 
resulting salt with alcohol. Colorless products were not 
obtained, due to some decomposition. Ligninsulphonic is 
probably dibasic, shows a high degree of ionization, and must 
be classed as a strong acid. On the other hand, solutions of 
barium ligninsulphonate even in great dilution are optically 
non-homogeneous, a part of the salt being, accordingly, in 
colloidal solution. In the salt solution, two systems of equilib- 
rium coexist: Colloid <=± Simple Molecules; Molecules <=± Ions; 
with increasing dilution both systems shift from left to right. 
The barium salt showed a molecular weight of 1110 from the 
freezing-point depression of an aqueous solution. Reduction 
of the salt with sodium and alcohol gave no definite products. 
With palladium and hydrogen 1.7470 grams of the barium salt 
took up 43.0 cubic centimeters of hydrogen, showing the presence 
of one reducible double bond to two atoms of sulphur, on the 
assumption that the reduction of an aldehyde group does not 
take place. 

By the action of sulphurous acid on spruce wood there is 
obtained a ligninsulphonic acid which, after purification by 
dialysis, behaves as a homogeneous substance of the formula 
C26H30O12S. 3 Its various reactions indicate the presence of the 
groupings C21H 15 2 . (S0 3 H) . (OH) 2 . (CH 2 OH) . (CHOH) . (CHO) .- 

1 R. H. McKee and G. Barsky, Paper Trade J., 74, 20 (1922), 46-48. 

2 F. Konig, Cellulosechemie, 2 (1921), 93-103, 105-113. 

3 C. Doree and L. Hall, J. Soc. Chem. Ind., 43 (1924), 257-263T. 



LIGNIN 83 

(OCH 3 ) 2 . The nitrated product does not behave as the nitro- 
derivative of a phenol, but of a non-benzenoid body of the nature 
of the terpenes. Oxidation with nitric acid of a concentration 
of 32 per cent gives oxalic acid and the acid C 2 oH 24 Oi 2 .(COOH) 6 .- 
(N0 2 )2- The continuance of the C 26 unit and the stability of a 
C 2 o nucleus throughout the various reactions lend support to 
Schrauth's 1 view that lignin is related to a reduced benzophenan- 
threne hydrocarbon. 

Cross and Bevan's Formula for Lignin. — The chlorine deriva- 
tive of lignin, C19H18CI4O9, called lignone chloride by Cross and 
Bevan, 2 is given a ketonic character. Its constitution is simi- 
lar to the chlorine derivatives of pyrogallol, namely, mairogallol, 

CO 



HC C(OH) 2 

and leucogallol, derived from 1 1 I , the general type, 

HC C(OH) 2 



CH 2 

and linked by oxygen bonds into three C6 complexes. 3 

Lignone chloride on heating gave a slight sublimate of 
chloroquinone and on reduction yielded trichloropyrogallol, the 
identifications being based on color reactions. Lignone chloride 
and the chlorine derivatives of pryogallol gave identical color 
reactions with sodium sulphite and ferric chloride. 

The above formula was subsequently elaborated to conform 
with additional experimental data on the behavior of the lignone 
complex and its relationship to a- and /3-cellulose. 4 

(A) (B) (C) (D) 

CO O O OH fa-cellulose 



HC CH-[CH 2 .C0] 2 -HC CH.CH. CH.CH 

HC CO CH3O.HC CH.OCH3 OH I ^-cellulose 

V \/ 

CH 2 CO 



1 W. Schrauth, Z. angew. Chem., 36 (1923), 149; Ber., 57 (1924), 854. 

2 C. F. Cross and E. J. Bevan, "Cellulose" (1895), p. 136. 

3 A. Hantzsch and K. Schniter, Ber., 20 (1897), 2033. 

4 "Researches" (1905-1914), p. 104; C. F. Cross, "Lectures on Celu- 

se" (1912). 



84 CHEMISTRY OF CELLULOSE AND WOOD 

In the benzenoid group A, chlorination takes place at the ethyl- 
ene and CH 2 positions to form a quinone chloride. With jute, 
combination with chlorine is one of substitution and addition, 
since the lignone chloride contains uniformly 26.7 per cent 
chlorine, and the chlorine as HC1 is one-half of that combined. 1 
The above structure would, however, give a lignin, as the alde- 
hyde, of the formula C20H22O10 and not C19H22O9, 2 by the elimina- 
tion of a molecule of water. The presence of a ketene group B 
explains the formation of acetic acid by hydrolysis, oxidation, 
and destructive distillation. The occurrence of a pyrone or 
hydropyrone group C is based on the production of acetone and 
maltol at elevated temperatures; while the reaction of ligno- 
cellulose with bisulphites and the formation of ligninsulphonates 
are due to the aldehyde group D. 

The formula proposed for lignin may be looked upon as a 
substituted diketide, which on hydrolysis at B would give the 
ketonic form of a dihydroxyphenylacetic acid. 

CO 



HC CH.CH2.COOH 

|| I + CH3.COOH + R. 

HC CO 



CH 



The evidence for the presence of a pyrone ring is not 
particularly strong. The small amount of acetone obtainable on 
destructive distillation could be formed by secondary decomposi- 
tion of acetic acid, while maltol is also obtained in the distillation 
of cellulose. 3 

Equally satisfactory theoretical formulae for the constituent 
groups of Cross and Bevan's lignin might be derived from ketene, 
CH 2 :CO, alone, owing to the extraordinary reactivity of this 
compound. 



1 "Cellulose" (1916), p. 104. 

2 "Researches" (1905-1910), p. 103. 

3 E. Erdmann and C. Schaefer, Ber., 43 (1910), 2398. 



LIGNIN 



85 



CH 

/\ 
Ketene-^Dehydracetic acid->OC C.CH 2 .COOH 

I I 

H 2 C CH 2 



(A) 



CO 

Dihydroxyphenylacetic acid 
(ketonic form) 
O 



Diacetylacetone 



H 2 



+2H 2 



CH3.C C.CH3 

II II 
HC CH 

\/ 
CO 

Dimethylpyrone 



CH 3 .HC 




O 



R.HoC.HC CH.CH 2 .R 



HO.HC CH.OH Meth y latlon CH3O.HC CH.OCH3 

\/ V 

CO CO 

(C) 

In fact, methods have been pointed out for the possible syntheses 
of the carbohydrates themselves from ketene. 1 

The presence of a carbonyl group in lignin cannot be deduced 
from its reaction with sulphites. As is well known, sulphurous 
acid combines readily with an ethylene group; e.g., citral reacts 
with sodium bisulphite to give the stable dihydrosulphonic 
derivative, C 9 Hi 7 .(S0 3 Na) 2 .CHO, that still retains the aldehyde 
group, since it reacts with phenylhydrazine. 2 Bucherer 3 found 
that some hydroxyl compounds combine readily with sulphurous 
acid. Resorcinol takes on three mols of H 2 S03 to form an ester 



1 J. N. Collie, J. Chem. Soc, 91 (1907), 1812. 

2 F. Tiemann and F. W. Semmler, Ber., 26 (1893), 2710; 31 (1898), 3314. 

3 H. Bucherer, Z. angew. Chem., 17 (1904), 1068; J. prakt. Chem., 69 
(1904), 49; 70 (1904), 345; Ber., 53 (1920), 1457-1459. 



86 CHEMISTRY OF CELLULOSE AND WOOD 

in which the S0 2 cannot be directly determined with iodine, but 
alkali splits it off readily. 

R.OH + NaHS0 3 = R.O.S0 2 .Na + H 2 0. 

He is of the opinion that lignin behaves like the tannins, forming 
an ester-like compound and not a sulphonic acid. According 
to Fuchs and Eisner, 1 sulphites react with the tautomeric form 
of the phenols to give ketone-bisulphite compounds. 

Phelps 2 found that "lignone" from sulphite liquor did not 
behave as a sulphonic acid derivative, since no phenols were 
obtained on fusion with alkali and there was no reaction with 
phosphorus pentachloride. He concluded that most of the 
sulphur is present as an acid sulphurous ether. 

Lignin Isolated with Alkalis. — The lignin in lignocelluloses 
is rendered soluble by fusion with alkalis or by heating with 
alkaline solutions under pressure; in the cold the solubility is 
limited. Rasped wood extracted with water, 5 per cent hydro- 
chloric acid, water, alcohol, ether, ammonia, and caustic soda 
(sp. gr. 1.10) was fused with potassium hydroxide at 185° by 
Lange. 3 Beech gave 12 per cent of lignin acids and 64 per cent 
of cellulose; oak, 14 per cent of lignin acids and 61 to 63 per cent 
cellulose; fir, 52 to 55 per cent cellulose. The alkaline filtrate 
from the cellulose gave on acidification lignin acids of the follow- 
ing composition: 





Beech, 
per cent 


Oak, 
per cent 


c 


61.50; 61.28 
5.32; 5.44 


60.78; 60.93 


H 


5.45; 5.40 



The lignin acids were partially soluble in alcohol. The soluble 
and insoluble acids were considered identical from the ultimate 
analyses and from the fact that the lignin acid insoluble in alco- 
hol, became soluble in alcohol to a large extent after solution 
in caustic soda, and reprecipitation with sulphuric acid. The 
lignin acids did not show the usual color reactions for lignin. 4 

'W. Fuchs and B. Elsner, Ber., 52 (1919), 2281; 53 (1920), 886; 64 
(1921), 245, 249; W. Fuchs and W. Stix, Ber., 55 (1922), 658. 

2 E. B. Phelps, U. S. Geol. Survey, Water-supply Paper, 226 (1909), p. 28. 

3 G. Lange, Z. physiol. Chem., 14 (1890), 15-30, 217-226. 

4 G. Lange, I.e., p. 219. 



LIGNIN 



87 





Beech, 
per cent 


Oak, 
per cent 


Fir, 
per cent 


Soluble in alcohol: 

C 


61.47 

5.48 

59.04 
5.37 


61.61 

5.47 

58.83 
5.15 


61.28 


H„ 


4.95 


Insoluble in alcohol: 

C 


60.51 


H 


5.22 







Aside from cellulose and lignin acids there were formed, during 
the fusion, formic acid, acetic acid, traces of higher fatty acids, 
oxalic acid, protocatechuic acid, catechol, ammonia, and traces 
of higher bases. 

Using Lange's method of fusion with caustic potash, Streeb 1 
obtained from spruce wood a lignin acid containing 64.59 per 
cent C and 5.35 per cent H. Acidification of the spent liquor 
from the manufacture of soda cellulose yielded a brown powder 
of variable composition: 62.94 to 65.50 per cent C; 5.11 to 5.39 
per cent H; and 5.78 to 5.98 per cent CH 3 . The lignin acids 
corresponded to the formula C 3 6H 44 0i4 or C36H40O12 and closely 
resembled the lignin acids of Lange. The acids were readily 
soluble in ammonia and caustic alkalis, difficultly soluble in 
alkali carbonates, insoluble in water, chloroform, benzene, and 
ether, and partially soluble in ethyl acetate and alcohol. 

Heuser 2 has stated that no oxalic acid was formed by the fusion 
of carbohydrate-free lignin with caustic potash at temperatures 
up to 270°. It was later 3 found that a 20 per cent yield of oxalic 
acid was obtainable when four parts of lignin were heated with 
50 parts of caustic potash at 280° for 40 minutes. The alkaline 
fusion of ligninsulphonic acid gives vanillic or protocatechuic 
acid, catechol, acetic acid, and traces of higher fatty acids. 4 

According to Honig and Fuchs, 5 the alkaline fusion of lignin- 
sulphonic acids gives no phenolic body other than protocate- 
chuic acid. The yield is 13 to 19 per cent, based on the organic 
portion. The sulphonic group of the barium salts employed 

1 E. Streeb, Diss. Gottingen (1892). 

2 E. Heuser, H. Roesch, and L. Gunkel, Cellulosechemie, 2 (1921), 113. 

3 E. Heuser and A. Winsvold, Cellulosechemie, 2 (1921), 113. 

4 K. H. A. Melander, J. Soc. Chem. Ind., 38 (1919), 625A. 

5 M. Honig and W. Fuchs, Monatsh., 40 (1919), 341. 



88 CHEMISTRY OF CELLULOSE AND WOOD 

was not completely removed until a temperature of 250 to 300 °C. 
was reached. It is, accordingly, believed that ligninsulphonic 
acids contain the carbon skeleton of protocatechuic acid. 

The three barium salts 1 obtained from sulphite liquor on boil- 
ing with a saturated aqueous solution of barium hydroxide are 
converted into different insoluble compounds and soluble sub- 
stances. 2 They are apparently identical with one another in 
composition and agree with the formula Ci 8 H 3 oOi SBa. The 

<OA TT 
nA 3 mT ) 

and has all the properties of a tannic acid of the catechu group. 
The salt in aqueous solution is almost quantitatively precipi- 
tated by hydrochloric acid and formaldehyde; it precipitates 
with gelatin and quinine hydrochloride, while hide powder 
absorbs about 70 per cent of the free acid. Iron and copper 
chlorides and lead acetate give greenish-gray, green, and white 
precipitates, respectively. 

Lignin on fusion with potassium hydroxide in the presence of 
air gave protocatechuic acid (16 to 19 per cent), catechol (1 to 

3 per cent), oxalic acid, and lignin acids. 3 Below 240° the main 
product of the fusion is lignin acid. When the fusion was con- 
ducted in an atmosphere of hydrogen or nitrogen, the yield of 
oxalic acid decreased and that of catechol increased to 9 per cent. 
When iron was present in the melt, the yield of catechol (23 per 
cent) increased, though the presence of protocatechuic acid 
could no longer be detected. Protocatechuic acid appears to 
be an intermediate product in the formation of catechol. In 
these results is seen additional evidence for the belief that lignin 
contains a benzenoid nucleus. 

Erdmann 4 on fusing purified fir wood with two parts of 
potassium hydroxide obtained acetic acid, catechol, and succinic 
acid. The source of the succinic acid (m.p. 180°) was attributed 

1 M. Honig and J. Spitzer, Monatsh., 39 (1918), 1. 

2 M. Honig and J. Spitzer, Monatsh., 41 (1920), 215-222. 

3 E. Heuser and A. Winsvold, Ber., 56 (1923), 902-909; Cellulosechemie, 

4 (1923), 49-58, 62-68. 

4 J. Erdmann, Ann. Suppl., 5 (1867), 228. The erroneous reference 
(Ann., 138 (1866), 1) cited by Lange has been repeatedly copied in the 
literature. 



LIGNIN 89 

to the sugar-producing portion of the wood. Bente 1 also 
obtained succinic acid by the alkaline fusion of wood. Lange 2 
was unable to confirm the findings of Erdmann with respect to the 
formation of succinic acid, though it has recently been identified 
in the alkaline fusion of "wood gum" from beech. 3 It has also 
been obtained by the oxidation of Willstatter's lignin with 
hydrogen peroxide. 4 

Beckmann 5 used various alkaline solutions for obtaining lignin 
from the straw of winter rye, and worked in the cold to avoid as 
much as possible a chemical change in the lignin. The best 
method consisted in treating the straw with a 2 per cent sodium 
hydroxide solution made from 600 cubic centimeters of 96 per 
cent alcohol, 20 grams of sodium hydroxide, and 400 cubic 
centimeters of water. The free alkali was neutralized with 
hydrochloric acid, the alcohol distilled off under reduced pressure, 
and the lignin precipitated with acid. Three extractions gave 7 
per cent of lignin, while the residual straw retained 15 per cent of 
lignin as determined with fuming hydrochloric acid. The rela- 
tionship of the easily to the difficultly soluble lignin remains to 
be determined. 

The lignin was readily soluble in basic solvents, such as 
pyridine, and acid solvents, such as the phenols. Various 
preparations gave on analysis : 

Per Cent 

C 61.80-63.01 

H 5.45-5.78 

CH 3 14.34-15.81 

Ratio C:CH 3 0, about 10:1 

The results agree with the formula C40H44O15 (mol. wt. 764.6). 
The sodium salt of the lignin contained somewhat less than 
two atoms of sodium. By the freezing-point method with phenol, 
the molecular weight varied from 762 to 1057, and by the boiling- 
point method with glacial acetic acid, from 768 to 891. Arylation 

1 F. Bente, Ber., 8 (1875), 478. 

2 G. Lange, Z. physiol. Chem., 14 (1890), 223. 

3 F. Konig, Cellulosechemie, 2 (1921), 118. 

4 E. Hagglund and C. B. Bjorkman, Biochem. Z., 147 (1924), 74. 

5 E. Beckmann, O. Liesche, and F. Lehmann, Z. angew. Chem., 34 
(1921), 285-288; Biochem. Z., 121 (1921), 293-310. 



90 



CHEMISTRY OF CELLULOSE AND WOOD 



established the presence of four hydroxyl groups, so that the 

/(OCH 3 )4 
formula may be represented by C 3 6H 2 80 7 \ 

X (OH) 4 

The effect of 1.5 per cent caustic soda on the lignins of wood and 
straw has been studied at room temperature, at the boiling point, 
and at 3, 6 ; and 9 atmospheres' pressure. 1 For woods, pressures 
of 6 and 9 atmospheres were necessary to remove all the lignin. 
The methoxyl content of the lignin obtained at 6 atmospheres' 
pressure was, in general, at a maximum. The total lignin 
obtained was always less than by the Willstatter method. Maple 
{Acer negundo) and pine (Pinus silvestris) lignins on reduction 
with hydriodic acid gave complex substances of the formulae 
C72H 65 0i 2 I and CiosHgsOigl, respectively. 

All natural lignins, according to Powell and Whittaker, 2 are 
derived from lignol, a polyhydroxy compound of the formula 
C 3 8H3o0 4 .(CO)2.(CHO).(OH) 9 . Lignins differ from each other in 
the number of methoxyl groups, which may be three to five. 
Data on the various lignins and their derivatives are given below : 



Sub- 
stance 
analyzed 



Determination 



Source of lignin 



Flax, 
per 
cent 



Larch, 
per 



Pine, 
per 
cent 



Spruce, 
per 
cent 



Ash, 
per 
cent 



Birch, 
per 
cent 



Poplar, 
per 

cent 



Lignin 



Acetyl- 
lignin 



Acetyl- 
methyl- 
lignin 



H 



-CHO 

-OCH 3 

-CO.CH3 

Number of methyl 

groups 

Number of acetyl 

groups 

Total groups 

-OCH3 

-CO.CH3 

Number of methyl 

groups 

Number of acetyl 

groups 

Total groups 



63.9 

5.8 

3.1 

11.8 

20.5 

4.0 

5.0 

9.0 

20.3 

11.1 



2.6 
9.1 



63.8 
5.2 

9.0 
23.0 

3.1 

5.8 

8.9 

19.9 

11.8 



2.7 
9.1 



63.4 
5.6 
2.9 

11.5 

18.9 

3.9 

4.5 

8.4 

20.4 

10.5 



2.4 
8.9 



64.0 
5.5 
3.1 

11.0 

19.4 

3.8 



23.1 

8.0 



7.2 
1.8 



63.2 
5.6 



13.3 
17.6 



4.5 
4.3 



20.0 
11.3 



2.6 
9.0 



63.2 

5.5 

3.2 

15.2 

14.5 

5.0 

3.4 

8.4 

22.4 

8.1 

7.0 

1.8 



63.3 

5.8 



12.6 
17.5 



4.3 
4.3 



22.7 
7.5 



7.0 



1.7 

8.7 



1 E. Beckmann, O. Liesche, and F. Lehmann, Biochem. Z., 139 (1923), 
491-508. 

2 W. J. Powell and H. Whittaker, J. Chem. Soc, 125 (1924), 357-364; 
127 (1925), 132-137. 



LIGNIN 91 

Examination of acetylated lignin showed that the total number of 
acetyl and methoxyl groups present represented nine hydroxyl 
groups, though the actual number may be eight to ten in lignin. 
The presence of an aldehydic group is shown by reduction of 
Fehling's solution; and since methyllignin reacts with three 
molecules of phenylhydrazine, two ketonic groups are probably 
present also. The insolubility of acetylated lignin in cold alkali 
indicates the absence of a carboxyl group. No information is 
given on the degree of acetylation of the lignin as it occurs in 
nature. 

The lignin was isolated by heating the raw material with 8 to 
12 per cent sodium hydroxide solution under pressure at 130 to 
160°, precipitating it with hydrochloric acid, and purifying by 
reprecipitation. It is assumed that by these treatments the 
lignin undergoes only slight change. This requires confirmation. 
It has been observed (Chap. XII) that a mild preliminary treat- 
ment of coniferous woods with either alkalis or acids renders the 
lignin much more difficult to remove than from the raw wood on 
cooking by the sulphite process. 

Lignin Isolated with Strong Mineral Acids. — The power of 
strong mineral acids to dissolve cellulose has long been known. 
Bechamp 1 employed strong sulphuric and hydrochloric acids, 
while in 1868 use was made of 72 per cent sulphuric acid for 
determining "ligneous cuticul." 2 Hydrochloric acid of a concen- 
tration of 40 to 42 per cent attacks wood rapidly. Spruce wood 
gave 30 per cent of lignin that was obtained purer than that by 
the use of sulphuric acid. 3 The lignin so isolated is much darker 
than the original lignocellulose, apparently as a result of dehydra- 
tion. In this way was explained the darkening action of strong 
sulphuric acid on the light-yellow lignin obtained as a residue 
from elder pith by the action of cuprammonium solution. 4 

Konig 5 has described several forms of lignin: Protolignin is 
rendered soluble by enzymes or water under a steam pressure 

1 A. Bechamp, Ann. chim. phys., 48 (1856), 463; Compt. rend., 42 (1856), 
1213. 

2 E. Fremy and A. Terreil, Bull. soc. chim., 9 (1868), 439. 

3 R. Willstatter and L. Zechmeister, Ber., 46 (1913), 2401. 

4 E. Fremy and Urbain, Compt. rend., 94 (1882), 109. 

5 J. Konig and E. Rump, "Chemie und Struktur der Pflanzen-Zellmem- 
bran" (1914), p. 85. 



92 



CHEMISTRY OF CELLULOSE AND WOOD 



of 2 to 3 atmospheres; hemilignin is dissolved by heating with 1 
to 3 per cent mineral acids at 1 to 3 atmospheres' pressure; 
ortholignin is partially dissolved by concentrated acids. The 
portion dissolving without coloration is colorless ortholignin, the 
residue, colored ortholignin. Unfortunately, most of the pains- 
taking analytical work was done on "crude fiber," so that the 
results are difficult to correlate. The lignin obtained by the use 
of 72 per cent sulphuric acid contained a small amount of sul- 
phuric acid that could not be removed by washing. Fir and 
beech lignin retained 0.045 per cent SO3. 1 Wood, purified by 
extraction with water and alcohol-benzene, after treatment with 
41 per cent hydrochloric by the Willstatter method, gave lignins 
with the following composition: 2 





Fir lignin from 


Beech lignin from 




Wood, 
per cent 


Crude fiber, 
per cent 


Wood, 
per cent 


Crude fiber, 
per cent 


Carbon 

Hydrogen 


62.62 
5.24 


66.15 
4.09 


60.95 
6.20 


63.72 

4.42 



Hagglund 3 prepared lignin from spruce, previously extracted 
with acetone and ether, by treating one part of wood with 10 
parts of 43 per cent hydrochloric acid and allowing to stand 15 
minutes. The yield was 28 per cent. The lignin gave the 
characteristic reactions with phloroglucinol, aniline sulphate, 
dimethyl-p-phenylenediamine, and ferricyanide, but not the 
Maule reaction. The lignin contained 3.5 per cent hygroscopic 
moisture, 1.63 per cent chlorine, 0.11 per cent nitrogen, and 0.95 
per cent ash. Correcting for these constituents, elementary 
analysis showed: C, 65.47 per cent; H, 5.47 per cent. Since 
the wood contained 4.98 per cent CH 3 0, the methoxyl content of 
the lignin should be 17.8 per cent; due to the action of the hydro- 
chloric acid, however, the isolated lignin contained but 14.39 
per cent CH 3 0. 

Since the lignin gave 3.69 per cent of furfural, it was apparently 
not free from pentosans. The lignin showed the following solu- 
bilities in various reagents: 

1 J. Konig and E. Rump, I.e., p. 57. 

2 J. Konig and E. Rump, I.e., p. 67. 

3 E. Hagglund, Arkiv Kemi, Mineral. Geol, 7, 8 (1918), 1-20. 



LIGNIN 



93 



Reagent 


Concen- 
tration, 
per cent 


Temperature, 

degrees 

Centigrade 


Time, 
hours 


Solu- 
bility, 
per cent 


Sodium bisulphite 

Calcium bisulphite 

Sodium hydroxide 

Sodium hydroxide 


5 

4 

10 

5 


140 
140 
100 
170 


5 
5 
2 
3 


12 
Traces 
15.2 

97.8 



The lignin prepared by Heuser, l by two treatments of spruce 
with 42 per cent hydrochloric acid, contained in the air-dry 
state 9.25 per cent moisture and 0.485 per cent ash. It had a 
copper number of 12.9, a methyoxyl content of 14.0 per cent, and 
gave no furfural on distillation with hydrochloric acid. 

The lignin obtained by treating spruce with 45 per cent hydro- 
chloric acid at 0° for 15 minutes contained 4 per cent of pentosan, 
identified as araban, so that Hagglund 2 considers the carbo- 
hydrate to be an integral part of the lignin. After removal of 
the pentosan from the lignin by hydrolysis, the copper number 
increased from 7.85 to 25.9. With 41 per cent hydrochloric acid, 
the lignin showed 6.64 per cent of pentosan after Yi hour and this 
decreased only slightly, to 6.38 per cent, after 12 hours. 3 Ktir- 
schner 4 likewise states that a lignin free from pentosan is obtained 
only after the prolonged action of cold fuming hydrochloric acid 
or hot dilute acid. After 3 hours' action, the lignin contained 
6 per cent of pentosans. Since xylose was obtainable from the 
ligninsulphonic acids from sulphite liquor, and from the lignin 
acids from soda liquor, he also concludes that some xylan is 
combined with the lignin. On the other hand, Powell and Whit- 
taker 5 found less than 1 per cent of pentosan in lignins isolated 
from wood with alkali, and this disappeared on purification. 

The lignin isolated by Ungar 6 by the action of hydrochloric 
acid (sp. gr. 1.204) on spruce for 25 hours contained almost all 

1 E. Heuser and C. Skioldebrand, Z. angew. Chem., 32 (1919), 41. 

2 E. Hagglund and C. J. Malm, Cellulosechemie, 4 (1923), 73; Ber., 
66 (1923), 1866. 

3 E. Hagglund, Cellulosechemie, 4 (1923), 84. 

4 K. Kurschner, Naturprodukte, 24 (1923), 36. 

5 W. J. Powell and H. Whittaker, J. Chem. Soc., 127 (1925), 132. 

6 E. Ungar, "Beitrage zur Kenntnis der Verholzten Faser," Diss. Buda- 
pest (1916), p. 63. 



94 CHEMISTRY OF CELLULOSE AND WOOD 

the methoxyl of the wood (15.2 per cent), about 2.5 per cent of 
chlorine, and gave 2.02 per cent furfural. 

Klason, 1 using 64 per cent sulphuric acid, obtained from spruce 
a lignin with C, 63.97 per cent; H, 5.32 per cent; and retaining 
1.70 per cent pentosan. 

Heuser 2 denies that pentosans are chemically combined with 
the lignin, their presence being due to incomplete hydrolysis. 
Xylan is more resistant to hydrolysis by fuming hydrochloric 
acid than by the hot dilute acid. 

Pine wood after exhaustion with acetone and ether contained 
22 to 26 per cent of lignin by Willstatter's method. By alternate 
treatment with cold concentrated hydrochloric acid and boiling 
dilute acid, more than half of the lignin was dissolved. The dis- 
solved portion contained arabinose, and a fermentable sugar, 
but no methylpentose. Since lignin free from carbohydrates is 
colored brown and gives a violet solution with strong hydro- 
chloric acid, while wood and its solution are colored green under 
the same treatment, it is concluded that a union exists between 
lignin and carbohydrates in the wood. 3 

This opinion is supported by the work of Miller and Swanson, 4 
who find that when wood is cooked by the sulphite process, the 
lignin behaves as if it were chemically combined with the easily 
hydrolyzable carbohydrates. Klason 5 has suggested that lignin 
is combined with the hemicelluloses by an acetal linkage. 

Lignin, like humin substances, in the opinion of Jonas, 6 is a 
condensed furan body and does not have a benzenoid structure. 
The dark-brown amorphous lignins isolated from the various 
conifers by the Willstatter method are identical in properties and 
composition. This lignin differs from the uncolored lignin in 
the plant, in that it is more highly polymerized, no longer con- 
tains acetyl groups, and is the methyl ester of a ketocarbonic 

1 P. Klason, Cellulosechemie, 4 (1923), 81; Svensk Pappers.-Tid., 26 
(1923), 319. 

2 E. Heuser, Cellulosechemie, 4 (1923), 77-78, 85. 

3 E. Hagglund and C. B. Bjorkman, Biochem. Z., 147 (1924), 74-78. 

4 R. N. Miller and W. H. Swanson, " Chemistry of the Sulphite Process. 
VIII. Studies of the Acid Hydrolysis of Wood," Paper read at 69th Meeting 
Am. Chem. Soc. (1925). 

5 P. Klason, Ber., 55 (1922), 455. 

6 K. G. Jonas, Z. angew, Chem., 34 (1921), 289, 373. 



LIGNIN 95 

acid. Its dark color is perhaps due to a carbonyl group, not 
originally present in the lignin, that could be formed by rearrange- 
ment of the primary hydroxyl group resulting from the splitting 
off of an acetyl group. In Willstatter lignin, methyl groups 
occur in combination both as esters and ethers. When treated 
with hydrogen bromide in glacial acetic acid, the lignin is com- 
pletely demethylated, becoming deep black and insoluble in all 
solvents. During demethylation, by rearrangement and split- 
ting off of a carbonyl group, another furan ring is formed. This is 
supported by the fact that, in distinction from the original 
product, the lignin will not pass over to an acid on heating with 
caustic soda under pressure; also it will no longer dissolve in 
boiling phenol. The furan structure of lignin is also supported by 
Marcusson. 1 

Willstatter 2 reduced the lignin from pine and beech with 
hydriodic acid (sp. gr. 1.7) and red phosphorus under pressure 
at 250°, obtaining a product separable into an acid portion and a 
mixture of solid and liquid hydrocarbons. The latter mixture 
contained 88.14 per cent C and 11.85 per cent H. The molecular 
weight varied from 167 for the lowest liquid fraction to 842 for 
the highest solid. No definite compounds were isolated. 
n-Hexyl iodide is not an intermediate product. Similar hydro- 
carbons were obtained by the reduction of cellulose, glucose, and 
other carbohydrates, indicating that there is a close structural 
relationship between lignin and the carbohydrates. During 
reduction the hydrocarbons are probably first converted to furan 
derivatives, while in lignin the furan structure preexists. The 
hydrocarbons appear to be a mixture of polycyclic, hydrogenated, 
five- and six-membered ring compounds in which the number of 
rings becomes greater with increasing molecular weight. In the 
light of the results of Willstatter, Schrauth 3 assumes that since 
carbohydrates give hydroxymethylfurfural with strong acids, 
lignin may be formed by the condensation of three molecules of 

!J. Marcusson, Z. angew. Chem., 34 (1921), 437; 35 (1922), 165; 36 
(1923), 42. 

2 R. Willstatter, L. Kalb, and G. von Miller, Ber., 55 (1922), 2637- 
2652. 

3 W. Schrauth, Z. angew. Chem., 36 (1923), 149, 571; Brennstoff-Chem., 
4 (1923), 161; W. Schrauth and K. Quasebarth, Ber., 57 (1924), 854. 



96 CHEMISTRY OF CELLULOSE AND WOOD 

5-oxymethylfurfural to give a benzophenanthrene derivative, 
CisHisOg, which by reduction in the plant goes over to Ci 8 Hi 8 6 . 
The derivative contains three — CO.CH 2 — groups which render 
the compound very reactive. Enolization produces a double 
bond and a phenolic hydroxyl group capable of methylation or 
esterification. 

Wood heated with formic or acetic acid containing 0.5 to 0.7 
per cent of sulphuric acid gives a ligneous product insoluble in 
water, soluble in formic and acetic acids, and partially soluble in 
alcohol, petroleum ether, and benzene. 1 

Gruss 2 obtained a lignin alcohol, C26H 46 Oio, by boiling beech 
wood with alcoholic hydrochloric acid. The alcohol is precipi- 
tated with water as a yellowish powder, m.p. 160°. It gives 
characteristic colorations with the various phenolic lignin 
reagents, but does not show the Schiff reaction for aldehydes. 
The alcohol is soluble in alkalis and reprecipitable with acetic 
acid; if the precipitate is boiled with 1.5 per cent sulphuric acid 
it is no longer soluble in alkali. The alcohol on oxidation with 
hydrogen peroxide at 60 to 70° gives a lignin acid whose copper 
salt, C7H14O5CU.5H2O, obtainable in a- and /3-modifications, has 
the noteworthy property of being crystalline. Analysis showed : 

Per Cent 

C 24.82 

H 7.20 

48.40 

Cu 19.10 

Wood dissolves readily in acetyl bromide, but slowly and 
incompletely in acetyl chloride. No action takes place unless 
moisture is present, so that the reaction is evidently one of 
hydrolysis and esterification. Zechmeister 3 found that when 2 
grams of finely divided wood were treated with 20 grams of 
acetyl bromide, only traces remained undissolved after standing 
one day. After standing 2.5 days, ice precipitates a compound, 
m.p. 60 to 70°, soluble in alcohol, ether, and benzene, and 
containing 25.5 to 26.1 per cent bromine. 

1 Akt.-Ges. fur Zellstoff-u. Papierfabr., G. P. 309551 (1916). 

2 J. Gruss, Ber. botan. Ges., 41 (1923), 48-52, 53-58. 

3 L. Zechmeister, Diss. Zurich (1913), p. 46; Ber., 56 (1923), 578. 



LIGNIN 



97 



oo 



OHO 



=HO 



OD' E HO 



O s HO 



So 






00 



CO fi a> 



«° TH' T+i 

g tH ^ GO 00 
lid 2 - » « 



- CO o o 
Sa 3! O © 

^ Tf CO CO 



» .5 .© -S 



S- © S- f- i- 



o P s 



§ £• 



<! o 



6 & S 8 « 



X K 

v o 



« 2 






fa o 

w CD 

% < 

o 



oM > M M m 



<! U < <) ■< « 
J H kJ J J O 

W W Q 



<! <! 

w n 



O O 
CU PL, 



3 



U U -Si 



«£ 3 s tf s 



coo 



o o 



C O* £ O* & 



"O T? 3 -3 






§1U 

— " — § ^ -^ 

» »» i; h ci 

S ; S ti p a a 

J3 ^ J J 3 3 

3 3 3 3 O O 

m m m m w m 



3 1 2 

3 © ft _ . 



P 3 X O 






5. °a. " ^ C Q- 

qqqqoooq o o q o o o q o o 
66666666 666666666 



b. 


OJ 


a 


CN 


!M 


O 


<N 


h- 


on 


a 


,_< 


O) 


CO 


<* 


1*1 


IO 




V 


no 


O 


::■. 










CM. 


CN 


Ol 


oi 


O) 


OJ 


IN 


CO 


X 


X 


X 


00 


as 


OS 


OS 


OS 


CT> 


C5 


OS 


OS 


OS 


OS 


OS 



98 CHEMISTRY OF CELLULOSE AND WOOD 

Five grams of dry beech or fir wood are immersed in 30 grams 
of acetyl bromide and let stand for 12 hours, whereby com- 
plete solution takes place. The amorphous granular precipitate, 
obtained by pouring on ice, is filtered off, washed with water, 
and then digested with alcohol to remove the cellulose derivative. 
The residual lignin derivative is obtained as a white amorphous 
powder by solution in acetone and precipitation with alcohol. 1 

Spruce lignin isolated by means of strong hydrochloric acid 
dissolves with much less rapidity than the wood, several months 
being required to render the greater portion soluble. 2 

Condensation Products of Lignin. — Phenols have the power to 
dissolve lignin at a temperature of about 200°. 3 By the addition 
of mineral acid to the phenol, the removal of the lignin is attained 
at a much lower temperature; 4 with 0.1 per cent sulphuric acid 
at 100° and with 1.0 per cent at 80°, the lignin can be removed in 
4 to 5 hours. 5 

Hochfelder 6 heated spruce, dried at 100°, with phenol at 180° 
for 48 hours. The dissolved lignin was obtained in two forms. 
Distilling off the phenol and extracting the residue with ether 
left "lignin substance a" as a brown, amorphous powder, while 
"lignin substance /3" was recovered from the ether as a resinous, 
syrupy mass. "Lignin substance a" has the formula C 2 2H 2 60 6 . 
It is insoluble in water, ammonia, soda, acetic acid, and benzene, 
and easily soluble in acetone, pyridine, and dilute caustic alkali. 
Carboxyl and carbonyl groups were absent, but the presence of 
three phenolic hydroxyl groups was shown by acetylation. 
Among the products resulting from the fusion of "lignin sub- 
stance a" with potassium hydroxide were identified acetic acid, 
traces of formic and isobutyric acids, salicylic acid, p-oxybenzoic 
acid, protocatechuic acid, oxalic acid (7.8 per cent), lignin 



1 P. Karrer and F. Widmer, Helvetica Chim. Acta., 4 (1921), 700; 
6 (1923), 817. 

2 E. Ungar, Diss. Budapest (1916), p. 77. 

3 F. A. Buhler, Papier-Ztg., 25 (1900), 3526; Chem. Ind. (1903), 138. 

4 R. Hartmuth, G. P. 326705; 328783 (1919). 

5 E. Legeler, Diss. Berlin (1921); Cellulosechemie, 4 (1923), 61; L. Kalb 
and V. Schoeller, Cellulosechemie, 4 (1923), 37. 

6 L. Hochfelder, "Beitrage zur Kenntnis der Ligninsubstanzen," 
Diss. Munchen (1915), 46 pp. 



LIGNIN 99 

acids (13.64 per cent), pyrogallol (?), catechol, and phenol 
(6.19 per cent). Owing to the condensation of phenol with 
lignin, phenol, salicylic acid, and some of the other products 
listed cannot be assumed, with Hochfelder, as originating from 
the lignin itself. The a- and /3-lignin substances contained, 
respectively, 1.83 and 0.59 per cent of pentosans. 

The solvent power of phenol is due to a condensation between 
the phenol and the lignin. 1 On treating Willstatter's lignin 
with boiling phenol, a condensation product is obtained that 
distills without decomposition between 230 to 240° at 10-milli- 
meters pressure. This is "phenol-lignin 6," a salve-like body. 
If the lignin is converted into an acid by heating with alkali 
under pressure before being condensed with phenol, there results 
"phenol-lignin a," a brown, amorphous, non-distillable product. 
Wood and Willstatter's lignin give identical phenol-lignins a 
and b. Phenol-lignin a has a certain resemblance to the phthal- 
eins; with alkalis it gives a dark-colored acid, that by reduction 
gives an almost colorless dihydroacid. The latter on oxidation 
passes at once into dark-colored "phenol-lignin a-acid." 

Buhler 2 proposed to use phenol for preparing puip from wood, 
but obviously the method is impracticable, due to the loss of 
phenol. According to Legeler, 3 in preparing 45 kilograms of 
pure cellulose from 100 kilograms of dry wood, 36 kilograms of 
phenol remain condensed with the non-cellulose constituents. 
Carbohydrates, including cellulose, under certain conditions give, 
with phenol and a mineral acid, bodies similar to phenol-lignin. 

The lignin residue obtained by the hydrolysis of wood, when 
condensed with phenol, gives a product having the character of a 
phenol. 4 It is difficultly soluble in alcohol and ether, but soluble 
in pyridine, phenol, acetone, and caustic soda. A similar prod- 
uct is obtained by heating wood with phenol and a small amount 
of hydrochloric acid at 90° for 1 hour. 

Norway spruce took up 1.33 to 2.32 per cent of aniline 
corresponding to 0.41 to 0.73 per cent of — CHO groups; and 0.64 



1 K. G. Jonas, Z. angew. Chem., 34 (1921), 289. 

2 F. A. Buhler, I.e. 

3 E. Legeler, Cellulosechemie, 4 (1923), 61. 

4 L. Kalb and V. Schoeller, G. P. 365287 (1920). 



100 CHEMISTRY OF CELLULOSE AND WOOD 

per cent hydroxylamine corresponding to 0.56 per cent — CHO. 
According to experimental conditions, the wood condensed with 
3.03 to 12.59 per cent of phloroglucinol dimethylether. l 

Phloroglucinol combines with the lignocelluloses in a definite 
ratio under specified conditions. 2 The reaction depends upon the 
amount of phloroglucinol and the concentration of the acid used. 
In the case of coniferous ground wood, using an amount of 0.5 
per cent phloroglucinol solution (in hydrochloric acid of sp. gr. 
1.06) equal to 10 per cent of phloroglucinol based on the wood, 
6.7 per cent is absorbed; with 1 per cent of phloroglucinol, the 
absorption is complete. If the wood after treatment with 10 
per cent of phloroglucinol solution is allowed to air dry, 8.9 per 
cent of phloroglucinol, C 6 H 6 03.2H 2 0, is combined due to a con- 
centration within the wood of both phloroglucinol and hydro- 
chloric acid. 3 Under the latter conditions, lignocellulose 
combines with 7.6 per cent pyrogallol. In the case of jute, acetyl- 
ation or chlorination did not alter the reaction with phloroglu- 
cinol. The same is true if the jute has first been digested in the 
cold for 6 days with 33 per cent hydrochloric acid with a loss in 
weight of 17 per cent, consisting mainly of the constituents yield- 
ing furfural. The phenols evidently condense with the lignin, 
since they are not removed by ordinary hydrolytic reagents, and 
no longer give the characteristic reactions with iron salts. 

The combination of a small amount of phloroglucinol with the 
lignocellulose produces a marked increase in the residue insoluble 
in 42 per cent hydrochloric acid. 4 Beech containing 30.1 per 
cent lignin, after combination with 5.1 per cent phloroglucinol, 
gave a ''lignin" residue of 76.3 per cent. This striking change is 
attributed to a new cleavage in the lignocellulose. 

Aromatic bases, such as dime thy 1-p-phenylenediamine, 
phenylhydrazine, and hydroxylamine, react with lignocellulose, 
the amount of nitrogen fixed being roughly proportional to the 
quantity of the base employed. While adsorption appears to be 



1 E. Ungar, Diss. Budapest (1916), p. 38. 

2 C. F. Cross, E. J. Bevan, and J. F. Briggs, Ber., 40 (1907), 3119- 
3126. 

3 C. F. Cross and E. J. Bevan, J. Soc. Dyers Colourists, 32 (1916), 135. 

4 C. F. Cross and C, Doree, "Researches," 4 (1922), p. 169. 



LIGNIN 101 

a large factor, a small portion of the base is evidently definitely 
combined with aldehydic groups in the lignocellulose. 1 

Lignin, obtained from straw by alkaline digestion, on heating 
with phenylhydrazine gave an amorphous phenylhydrazone, 
C58H57N6O10, soluble in alcohol, acetone, and tetrachlorethane ; 
with the atter solvent, transparent varnish-like coatings are 
obtainable. 2 When 1.5 parts of lignin were heated for 2 hours 
with one part of nitrosodimethylaniline, one part of concen- 
trated hydrochloric acid, and 50 parts of alcohol, there was 
obtained lignocyanine, C101H103N16O26, resembling gallocyanine. 
It dyes silk and mordanted cotton a brownish violet. The 
heating of equal parts of lignin, aniline, and oxalic acid at 180° 
for 30 minutes gave a condensation product containing 4.7 per 
cent nitrogen. This corresponds with a compound, C 5 8H 6 oN 3 Oio, 
containing three molecules of aniline to one of lignin. 3 A brown 
sulphur dye, C36H49S6O21, is obtained on fusion with sodium 
sulphide. In its various reactions straw cellulose acts as a 
trivalent substance, hence containing three aldehydic or ketonic 
groups. Further support for its aldehydic nature is found in 
the fact that an alkaline solution of straw lignin used on hides 
results in a soft-brown leather, without increase in weight, 
while a skin treated with an alkaline solution of sulphite liquor 
increased in weight about 75 per cent but was not tanned, and 
became hard and brittle after drying. None of the condensation 
products are definite substances, and their formation throws little 
light on the structure of lignin. 

Action of Halogens. — Payen, 4 in 1839, observed that chlorine 
had both a bleaching and a yellowing action on wood. Von 
Baumhauer 5 also used chlorine gas in investigating the composi- 
tion of fruit pits. A quantitative method for the determination 
of cellulose was developed by Fremy and Terreil 6 through the 
observation that after treatment with chlorine the incrusting 

1 C. F. Cross, E. J. Bevan, and J. F. Briggs, Ber., 40 (1907), 3119; 
C. F. Cross and E. J. Bevan, J. Soc. Dyers Colourists, 32 (1916), 135. 

2 F. Paschke, Cellulosechemie, 3 (1922), 19-21. 

3 F. Paschke, Cellulosechemie, 4 (1923), 31-32. 

4 A. Payen, Compt. rend., 9 (1839), 149. 

6 E. H. von Baumhauer, J. prakt. Chem., 32 (1844), 212. 
6 E. Fremy and A. Terreil, Bull. soc. chim., 9 (1868), 439. 



102 



CHEMISTRY OF CELLULOSE AND WOOD 



materials became soluble in alkalis. Bevan and Cross 1 boiled 
the chlorinated fibers in sodium sulphite solution, a modification 
largely retained today. Sodium bisulphite is much less efficient 
than sodium sulphite for dissolving the chlorinated lignin, 
showing that an alkaline reagent is required. 2 

Cross and Bevan 3 by the direct chlorination of jute obtained 
a lignin chloride, Ci 9 Hi 8 9 Cl4, with 26.8 per cent chlorine. While 
dissolving most readily in alkaline solutions, lignin chloride also 
dissolves in glacial acetic acid and alcohol. In the case of jute, the 
weight of chlorine reacting is 16 to 17 per cent of the weight of the 
fiber, 8.0 to 8.5 per cent appearing as hydrochloric acid; hence, 
the reaction is one of simple substitution. Dry lignocellulose 
does not react with chlorine even when heated with it at 60 to 
80°. In the case of woods, the amount of hydrochloric acid 
formed is considerably greater. The following results are based 
on wood previously exhausted with 1 per cent caustic soda. 4 





Pine, 
per cent 


Beech, 
per cent 


Sycamore, 
per cent 


Birch, 
per cent 


Combined chlorine 

Acidity (as HC1) 


7.5 
23.5 


7.5 
19.5 


9.0 
21.0 


7.0 
15.0 



One gram of jute 5 absorbed 60 cubic centimeters of chlorine at 
18°. 

The furfural-producing carbohydrates are not changed by 
chlorination, since the yield of furfural from a fiber is the same 
before and after chlorination. 6 Hagglund 7 obtained furfural 
from spruce lignin isolated with strong hydrochloric acid, but 
after chlorination of the lignin no furfural was obtainable. The 
chlorine content of the lignin was not uniform, varying from 
39.7 to 46.2 per cent. 



1 E. J. Bevan and C. F. Cross, Chem. News, 42 (1880), 80. 

2 A. W. Schorger, J. Ind. Eng. Chem., 9 (1917), 561. 

3 "Cellulose" (1895), p. 102; cf. J. Chem. Soc., 55 (1889), 199. 
* C. F. Cross and E. J. Bevan, "Cellulose" (1895), p. 195. 

5 E. J. Bevan and C. F. Cross, Chem. News, 58 (1888), 215. 

6 "Cellulose" (1916), p. 136. 

7 E. Hagglund, Arkiv Kemi, Mineral. Geol., 7 (1918), 18. 



LIGNIN 



103 



Heuser and Sieber, 1 in studying the action of chlorine on 
spruce wood, found (Fig. 3) that up to the point (2 hours) where 
all the lignin was removed, 31.1 per cent of chlorine was obtained 
as hydrochloric acid, while only 9.47 per cent had combined with 
the lignin; accordingly, the action of chlorine on wood lignin at 
least is mainly one of oxidation. Chlorination when extended 
to 22 hours resulted in a residue of oxycellulose, though the loss 
in weight was comparatively slight. As long as lignin remained 
there was practically no formation of oxycellulose. The lignin 
chloride gave on analysis: C = 47.03 per cent; H = 4.59 per 

50 100 



-£45 90 
I 40 "I 80 

^35 |_70 
I 30 -£ GO 

£ 25 | 50 

^ 20 ^ 40 
~ 15^30 



_o 

I 10 

■i 5 





— >-J 




To Left - 


Ligrrocellufose and Ce/Jufose 










To Rigfrr-Ce/Iufose and Oxyce//u/ose 




















\ 1 1 1 
Hydrochloric Acta 


f 










<~~ 












"TTT 


t — 










; 


' 




— •"■ 


— •— 


























/ 
/ 


































/ 

/ 


































/ 


































i 


































i 



































Fig. 3. 



20 

10 


I 2 3 4 5 G 7 8 9 10 I! 12 13 14 15 16 17 

Ti'me o*f Reac+ion in Hours 
-Progressive action of chlorine on spruce wood. (After Heuser and 
Sieber.) 



cent; CI = 22.68 per cent. Lignin chloride is not considered to 
be a uniform compound nor to have a cyclic structure, since it 
could not be reduced to a derivative of pyrogallol, as claimed by 
Cross and Bevan. It did not give a trace of furfural on distilla- 
tion with 12 per cent hydrochloric acid. Fusion with caustic 
alkali gave large amounts of oxalic acid. 

Jonas 2 claims that if Willstatter's lignin is carefully chlorinated 
with good cooling, one and the same chloride is always obtained. 
Chlorination is accompanied by a partial oxidation, the light- 
yellow lignin chloride obtained being soluble in alkali, alcohol, 
and acetone. 

1 E. Heuser and R. Sieber, Z. angew. Chem., 26 (1913), 801-806. 

2 K. G. Jonas, Z. angew. Chem., 34 (1921), 289. 



104 CHEMISTRY OF CELLULOSE AND WOOD 

Straw lignin, at room temperature, gave with sulphuryl 
chloride a pale-brown amorphous compound, C37H42S3CI3O12, 
free from methoxyl groups. 1 When the reaction was performed 
by heating in a sealed tube for 1 hour at 100°, there was obtained 
a yellow resinous body, C36H42CI11O10, free from sulphur but 
containing 38.22 per cent chlorine. A lignin derivative, 
C38H46CI5O15, containing only 19.18 per cent chlorine, resulted 
from the action of two parts of phosphorous pentachloride in 
tetrachlorethane. 

Lignin is profoundly chlorinated when boiled for 3 hours with 
antimony pentachloride containing a small amount of iodine. 
Perchlorethane and, apparently, hexachlorobenzene are among 
the products formed. 2 

Bromine behaves towards lignin like chlorine, though much 
less energetically. Muller 3 removed the lignin from plant 
fibers by alternate treatment with bromine water and ammonia. 
A greater combination with bromine is obtained if the lignocellu- 
lose is first dissolved in zinc chloride-hydrochloric acid, bromine 
added, and, after standing about 16 hours, precipitating with 
water. 4 In this way the non-cellulose portion of esparto gave 
the compounds Ci7Hi4Br 4 6 , Ci 5 Hi 2 Br 4 05, and C27H 2 8Br 4 Oio. 

Iodine appears to be without chemical reaction on the ligno- 
celluloses, forming as with cellulose 4 only adsorption compounds. 

Lignin isolated from flax and wood with caustic soda, when 
suspended in carbon tetrachloride and treated with chlorine, 
gives a brick-red solid having the formula C4oH 2 o08Cli2(OCH 3 )2.- 
(OH) 5 .CHO. Regenerated with acids from its solution in cold 
alkali, it has only 6 chlorine atoms. The bromolignin is a 
dark-red solid similar to chlorolignin. 5 

Acetyl and Formyl Groups. — The ease with which acetic and 
formic acids may be obtained by the use of hydrolytic agents 
indicates that acetyl and formyl groups are present in the ligno- 

1 F. Paschke, Cellulosechemie, 3 (1922), 19. 

2 H. Tropsch, Ges. Abhandl. Kenntnis. Kohle, 6 (1921), 301-302. 

3 Hugo Muller, ''Die Pflanzenfaser." In A. W. Hoffmann's "Bericht 
uber die Entwichlung d. Chem. Industrie," 2 (1877), 27. 

4 C. F. Cross and E. J. Bevan, "Cellulose" (1916), p. 138. 

5 W. J. Powell and H. Whittaker, J. Chem. Soc., 125 (1924), 357; 
127 (1925), 132. 



LIGNIN 



105 



celluloses as a part of the lignin. Cross and Be van 1 obtained 7 
to 8 per cent of acetic acid by treating beech wood with dilute 
alkalis. Acetic acid is also obtainable by digesting woods with 
dilute sulphuric acid at 60 to 100°. Larger yields are obtainable 
by treating the wood with cold concentrated sulphuric acid, dilut- 
ing, and distilling. 

Cross 2 and his associates found that on distilling celluloses 
with sulphuric acid the amount of volatile acids, particularly 
formic, increased greatly with the strength of the sulphuric 
acid. Barley straw gave 1.7 and 14.8 per cent of volatile acids, 
calculated as acetic acid, with 10 and 50 per cent sulphuric acid, 
respectively. The amounts of acid given by various celluloses 
on distilling with 40 to 50 per cent sulphuric acid were : 



Formic acid, 
per cent 



Swedish filter paper. 
Esparto cellulose .... 

Bleached cotton 

Raw cotton 

Jute cellulose 

Beech-wood cellulose 




17.2 

16.6 

13.2 

9.4 

22.7 
14.6 



If the presence of acetyl groups is a characteristic feature of 
lignification, there must be a marked difference between the 
lignins of hardwoods and conifers. On boiling with 2.5 per cent 
sulphuric acid for 3 hours the hardwoods give 4.30 to 5.79 per 
cent acetic acid and the conifers 0.71 to 1.79 per cent. 3 A slight 
amount of formic acid is formed simultaneously, but it is ques- 
tionable if formyl groups are actually present, since it is 
characteristic of hexose carbohydrates to decompose to lsevulinic 
and formic acids on boiling with dilute acids. 

Data have been obtained by Dore 4 indicating that acetyl 
groups may sometimes be associated with the cellulose. Due 
to saponification by the strong acid, only a small amount of 



1 "Cellulose" (1895), p. 191. 

2 C. F. Cross et al, Ber., 28 (1895), 1940. 

3 A. W. Schorger, J. Ind. Eng. Chem., 9 (1917), 556. 

4 W. H. Dore, J. Ind. Eng. Chem., 12 (1920), 475. 



106 CHEMISTRY OF CELLULOSE AND WOOD 

acetic acid remains in the lignin. As would be expected, the acetyl 
groups are completely removed by digestion in the cold with 5 
per cent sodium hydroxide. 1 The acetic acid should exist in 
the solution as sodium acetate and not as "substances yielding 
acetic acid on hydrolysis." 

Distribution of Acetio Acid in Redwood 

Acetic Acid, 
Per Cent 

Original wood . 83 

Cellulose (by chlorination) . 55 

Lignin (by 72 per cent H 2 S0 4 ) .09 

Cross and Bevan 2 are of the opinion that the acetic acid is 
derived from ketene derivatives rather than from acetyl groups. 

The formation of acetic and formic acids by the acid hydrolysis 
of lignocelluloses is definitely attributed to the presence of acetyl 
and formyl groups. 3 Beech wood, on treatment in an autoclave 
for 3 hours with 1 per cent sulphuric acid, gave 5.2 per cent of 
acid, as acetic acid. Formic acid was present, but higher fatty 
acids and laevulinic acid were not detected. Spruce ground wood 
gave 1.06 per cent of acetic acid and 0.21 per cent of formic acid, 
the ratio being 1:5. Spruce wood that had been hydrolyzed 
twice with acid under pressure gave, after washing and drying, 
2.27 per cent of acid on destructive distillation. The original 
wood gave only a slightly greater yield of acid, namely, 2.43 per 
cent, on destructive distillation. 

Guijo, a Philippine wood having a marked corrosive action 
on metal fastenings, yielded 0.2 per cent of acetic and formic 
acids on leaching with cold water. 4 This acidity may be due to a 
gum such as that produced by Cochlospermum gossypium, a 
small deciduous tree growing in India, which has the property 
of giving off acetic acid on exposure to moist air. It yields 14.4 
per cent of acetic acid on hydrolysis with sulphuric acid. 5 Cross 

1 W. H. Dore, J. Ind. Eng. Chem., 12 (1920), 986. 

2 C. F. Cross and E. J. Bevan, "Researches" (1905-1910), p. 103; 
(1910-1921), p. 164. 

3 W. E. Cross and B. Tollens, J. Landw., 59 (1911), 185-196; Ber.. 
43 (1910), 1526-1528. 

4 A. W. Schorger, J. Ind. Eng. Chem., 9 (1917), 561. 

5 H. H. Robinson, J. Chem. Soc., 89 (1906), 1496. 



LIGNIN 107 

and Tollens 1 found that "wood gum" gave no acetic acid on 
hydrolysis. 

The acetyl content of wood was determined by Pringsheim 
and Magnus 2 by heating it with eight times its weight of 3.5 
per cent sodium hydroxide solution at 6 atmospheres for 6 
hours. Pine wood gave 4.2 per cent acetic acid. Repetition 
of the treatment for 3 hours gave an additional 1.5 per cent, 
while a third treatment gave only a trace. In comparison with 
5.7 per cent of acetic acid from pine, hornbeam gave 10.5 per 
cent of acetic acid. The acetyl groups of Willstatter's lignin 
are completely removed by the acid treatment, but can be 
restored by acetylation, the product becoming appreciably 
lighter in color by the operation. The lignin of coniferous wood 
contained 19.85 per cent acetic acid and the lignin of hornbeam 
37.8 per cent, as determined by acetylation and subsequent 
saponification. On the basis of 23 per cent lignin (Willstatter) 
and 10.5 per cent acetic acid by the alkaline hydrolysis of horn- 
beam, and 30 per cent lignin and 5.7 per cent of acetic acid from 
pine, the acetic acid content of the acetylated lignin is 5 to 7 
per cent higher than in the original lignin. It is claimed by the 
authors that the acetyl groups of wood cannot be completely 
saponified by heating for 3 hours with 2.5 per cent sulphuric acid 
and that the increased yield of acetic acid by alkaline saponifi- 
cation is due in only small part to the cellulose. This is scarcely 
correct. Their data show that a wood cellulose freed from lignin 
by the chlorination method, and from pentosans by glycerol- 
sulphuric acid mixture, gave 3.05 per cent acetic acid when boiled 
with caustic soda, and 4.6 per cent acetic acid with alkali at 4 
atmospheres pressure. It has been quite correctly objected that 
probably only about one-third of the acetic acid obtained by the 
Pringsheim method has its source in the lignin. 3 

When finely divided wood is extracted with an alkaline solu- 
tion at ordinary temperature or by gentle heat at atmospheric 
pressure, fatty acids are obtained, consisting mainly of acetic 
acid, with 1 to 3 per cent of formic acid and a trace of butyric 

1 L.e. 

2 H. Pringsheim and H. Magnus, Z. physiol. Chem., 105 (1919), 179- 
186; Z. angew. Chem., 33 (1920), 56. 

3 C. Schwalbe and E. Becker, Z. angew. Chem., 33 (1920), 225. 



108 



CHEMISTRY OF CELLULOSE AND WOOD 



acid. Pine wood gives 1 to 2 per cent of acids, and beech 5 to 6 
per cent. 1 

Methoxyl Groups. — The methyl groups in wood occur largely 
as ethers, but a small part apparently as an ester. In an 
extensive investigation of the methyl content of wood Benedict 
and Bamberger, 2 using the Zeisel method, showed that the 
presence of CH 3 groups was characteristic of lignification. 
Cotton and filter paper gave no methoxyl and sulphite cellulose 
but 0.70 per cent. In the following table the original methyl 
numbers have been recalculated to methoxyl. 



Species 



CH 3 0, 
per cent 



Maple (Acer pseudoplatanus L.) . . . . 
Locust (Robinia pseudacacia L.) 

Birch (Betula alba L.) 

Pear (Pyrus communis L.) 

Oak (Quercus pedunculata Ehrh.) . . 

Alder (Alnus glutinosa Gart.) 

Ash (Fraxinus excelsior L.) 

Ash (Fraxinus excelsior L.) 

Spruce (Abies excelsa D. C.) 

Spruce (Abies excelsa D. C.) 

Spruce (Abies excelsa D. C.) 

Pine (Pinus sylvestris L.) 

Larch (Larix europea, D. C.) 

Fir (Abies pectinata D. C.) 

Linden (Tilia parvifolia Ehrh.) 
Mahogany (Swietenia mahogoni L.) 

Walnut (Juglans regia L.) 

Poplar (Populus alba L.) 

Beech (Fagus silvatica L.) 

Elm (Ulmus campestris L.) 

Willow (Salix alba L.) 

Hornbeam (Carpinus betulus L.) . . . 



Trunk 

Branch 

Trunk (3 years old) 

Trunk 

Trunk 

Trunk 

Trunk 

Branch 

Trunk 

Trunk (center) 

Trunk (periphery) 

Trunk 

Trunk 

Trunk 

Trunk 

Trunk 

Trunk 

Trunk 

Trunk 

Trunk 

Trunk 

Trunk 



6.30 
4.88 
5.29 
6.61 

5.42-5.87 
5.95 
5.58 
6.22 

4.43-4.92 
5.33 
4.78 
4.63 

4.10-5.52 
5.05 
5.27 
5.48 

4.68-5.54 
5.33 

5.40-6.22 
6.01 
4.76 
5.48 



The authors conclude that, in general, the branch is richer in 
lignin than the trunk and that in the latter the center is more 
lignified than the younger annual growths. 



1 J. Ederer, G. P. 272036, (1913). 

2 R. Benedict and M. Bamberger, Monatsh., 11 (1890), 260-267. 



LIGNIN 



109 



A similar series of methoxyl determinations was made on 
American species by Wheeler. 1 

CH 3 o, 

Species Per Cent 

Persimmon (Diospyros virginiana L.) 4 . 02 

Umbrella tree (Magnolia tripetala L.) 5 . 29 

Sassafras (Sassafras sassafras (L.) Karst.) 5.03 

Chinquapin (Castanea pumila (L.) Mill.) 4.45 

Buttonwood (Platanus occidentalis L.) 4.59 

Witch hazel (Hamamelis virginiana L.) 5 . 50 

Pignut hickory (Hicoria glabra Britt.) 4 . 78 

Sweet gum (Liquidambar styraciflua L.) 4 . 61 

Dogwood (Cornus florida L.) 4 . 88 

Honey locust (Gleditsia triacanthos L.) 5 . 09 

Crabwood (Gymnanthes lucida Swartz) 6 . 08 

The methoxyl content of the angiosperms will average 
somewhat higher than the gymnosperms, though the latter are 
higher in lignin. 

Cellulose isolated by the chlorine method gives a methoxyl 
reaction. This is considered by some investigators as due to 
residual lignin, while Cross and Bevan 2 believe that CH 3 groups 
are characteristic of ^-cellulose. Jute cellulose contains about 
1.2 per cent methoxyl 3 in comparison with 4.6 per cent for jute. 
Since /3-cellulose forms about 20 per cent of the jute cellulose, 
the methoxyl content of the 0-cellulose would be 6.0 per cent. 
The methoxyl values of celluloses isolated by various methods 
are: 4 



Method 


Oak, per cent 


Fir 


, per cent 


Tollens 


0.28 
0.49 
0.72 
2.50 
1.00 




0.43 


Konig 


0.68 


Cross and Bevan 




1.45 


F. Schulze 

H. Muller 




2.59 
1.23 







1 A. S. Wheeler, Ber. 38 (1905) 2168-2169. 

2 C. F. Cross and E. J. Bevan, "Cellulose" (1916), pp. 93-94. 

3 L.c, p. 158. 

4 J. Konig and F. Huhn, " Bestimmung der Zellulose in Holzarten" 
(1912), p. 47. 



110 



CHEMISTRY OF CELLULOSE AND WOOD 



Spruce lignin isolated by the Willstatter method contained 
about 14 per cent CH 3 0, some being lost by hydrolysis. Lignin 
acids prepared by Lange's method of alkaline fusion gave by the 
Zeisel method 13.11 per cent CH 3 (13.55 per cent CHsOH). 1 

Heuser 2 methylated lignin isolated from wood with 
hydrochloric acid (sp. gr. 1.2) by suspending 5 grams in 100 
cubic centimeters of 10 per (tent sodium hydroxide, and treating 
at 60° with 30 grams of dimethylsulphate. The original CH 3 
content was thereby increased from 14.65 per cent to 20.65 per 
cent with a yield of 96 per cent. By repeated methylation the 
CH 3 content was raised to 26.29 per cent. Removal of the 
methoxyl by repeated heating in a sealed tube with hydrochloric 
acid gave a residue containing: C, 67.45 to 69.30 per cent; H, 
4.16 to 4.50 per cent. Three methylations of this residue gave 
a methyllignin with 5.79 per cent CH 3 0. Lignin rendered soluble 
by heating with caustic soda in an autoclave at 170° gave on 
methylation a product with 24.7 per cent CH 3 0. Another 
sample of lignin containing 14.15 per cent methoxyl gave on 
repeated methylation a lignin with 26.05 per cent CH 3 0. The 
methoxyl content of ligninsulphonic acid was in the same way 
increased from 13.07 to 25.43 per cent. 3 

The methoxyl content of lignins isolated by the use of strong 
mineral acids varies greatly between species, but there is more 
uniformity in the conifers than in the hardwoods. Lignin 
isolated from fir and beech with 72 per cent sulphuric acid con- 
tained 13.90 and 16.64 per cent methoxyl, respectively. 4 
Beckmann 5 found the following: 



Picea 
excelsa, 
per cent 



Pinus 
silvestris, 
per cent 



Acer 
negundo, 
per cent 



Fagus 
silvatica, 
per cent 



Lignin (Willstatter) 
CH3O therein 



29.90 
15.77 



29.83 
15.70 



26.14 
16.46 



22.36 
19.49 



1 T. von Fellenberg, Biochem. Z., 85 (1918), 79. 

2 E. Heuser, R. Schmitt, and L. Gun ke l, 'Cellulosechemie, 2 (1921), 
81-86. 

3 E. Heuser and S. Samuelsen, Cellulosechemie, 3 (1922), 78. 

4 J. Konig and E. Rump, I.e., p. 63. 

5 E. Beckmann et al, Biochem. Z., 139 (1923), 491. 



LIGNIN 



111 



The extensive series of analyses by Ritter and Fleck 1 show a 
great variation in the lignin content of woods and the methoxyl 
content of the lignins isolated by the use of 72 per cent sulphuric 
acid. Their results are given in the table below. 

Methoxyl in Lignin from Sapwood and Heartwood 





Lignin in 


CH 3 in lignin 


Species 


Sapwood, 
per cent 


Heartwood, 
per cent 


Sapwood, 
per cent 


Heartwood, 
per cent 


White ash 


f 26 . 95 
\ 27.39 
\ 23.08 
\ 23.86 

21.87 
f 25.87 
\ 26.64 
\ 35.01 
\ 35.31 

26.52 

29.03 
f 29.85 
\ 32 . 14 

34.73 


27.39 
28.38 
22.19 
23.69 
22.85 
25.68 
25.94 
33.06 
32.27 
26.14 
28.73 
31.39 
32.42 
33.67 


17.4 
20.0 
25.2 
24.7 
25.4 
20.4 
19.8 
12.4 
14.0 
15.7 
15.1 
17.1 
16.2 
17.1 


19.5 


Yellow poplar 

Black hickory 

Red alder 


18.3 
26.4 
25.4 
25.3 
20.1 


Bald cypress 


20.3 
11.9 


Eastern white pine .... 
Yellow cedar 


12.6 
17.6 
16.7 


Southern white cedar. . 
Incense cedar 


15.9 
15.7 
18.4 







It had been found 2 that the isolated lignins of redwood and 
live oak contained approximately 18 per cent of methoxyl, but 
the above analyses show variations from 11.9 per cent in the 
case of bald cypress to 26.4 per cent in yellow poplar. 

With conifers, 10 to 21 per cent of the methoxyl, and with 
hardwoods, 21 to 38 per cent of the methoxyl is not recovered 
in the isolated lignin. 3 While part of the loss of methoxyl is 
due to subsequent boiling of the diluted acid, as recommended by 
Mahood, 4 the results indicate that if the methoxyl is wholly 
associated with the lignin, the lignins vary in chemical composition 



1 G. J. Ritter and L. C. Fleck, J. Ind. Eng. Chem., 15 (1923), 1055. 

2 G. J. Ritter and L. C. Fleck, J. Ind. Eng. Chem., 14 (1922), 1050. 

3 G. J. Ritter, J. Ind. Eng. Chem., 15 (1923), 1264. 

4 S. A. Mahood and D. E. Cable, /. Ind. Eng. Chem., 14 (1922), 933. 



112 CHEMISTRY OF CELLULOSE AND WOOD 

or in the mode of the methoxyl linkage. Dore 1 found that in 
redwood the methoxyl groups were entirely associated with the 
lignin, while in live oak (Quercus agrifolia), containing 5.80 per 
cent CH 3 0, 2.08 per cent of the methoxyl appeared to be differ- 
ently associated or very loosely combined with the lignin. 2 

Western yellow pine and white oak when repeatedly heated 
with dilute alkali under pressure gave 63 per cent of their meth- 
oxyl as methyl alcohol. Using the Zeisel method, the remaining 
methoxyl could be accounted for in the residue from white oak, 
but not in the case of western yellow pine. 3 These results are 
in interesting agreement with those obtained by destructive 
distillation, wherein 62 to 63 per cent of the methoxyl was 
recovered as methyl alcohol. 4 This suggests that the same 
methoxyl groups in the wood yield the methyl alcohol by alkaline 
hydrolysis, and the methyl alcohol and methoxyl groups in 
products of destructive distillation. 

If the methoxyl in wood is all combined with the lignin, and 
in the same way, the ratio of lignin to methoxyl in the residue 
from wood subjected to varying periods of chlorination should 
be the same as for the original wood. This appears to hold for 
basswood but not for tanbark oak (Quercus densiflora) and incense 
cedar. 5 

Von Fellenberg 6 believes that the methyl alcohol formed by 
treating wood with dilute alkali is derived from pectin, and that 
obtained with strong sulphuric acid from lignin. Support for 
Konig's view that there are two or more forms of lignin is found 
in the fact that some methyl alcohol is formed on heating with 
18 per cent sulphuric acid, though no sharp line of hydrolysis 
is obtained on increasing the strength of the acid to 72 per cent. 
The distribution of the methyl alcohol in an ash tree 5 years old 
is given in the following table. The methyl alcohol in lignin I 
was determined by heating the wood with 40 per cent sulphuric 
acid, while 72 per cent acid was used for lignin II. 

1 W. H. Dore, J. Ind. Eng. Chem., 12 (1920), 475. 

2 W. H. Dore, J. Ind. Eng. Chem., 12 (1920), 986. 

3 G. J. Ritter, J. Ind. Eng. Chem., 15 (1923), 1264. 

4 L. F. Hawley and S. S. Aiyar, J. Ind. Eng. Chem., 14 (1922), 1056. 

5 G. J. Ritter, J. Ind. Eng. Chem., 15 (1923), 1264. 

6 T. von Fellenberg, Biochem. Z., 85 (1918), 45-161. 



LIGNIN 



113 



Per cent methyl alcohol in dry substance 



Part of plant 



Wood 



Bark 



13 



Root hairs 

Main root 

Shoot 5 years old 

Shoot 4 years old 

Shoot 3 years old 

Shoot 2 years old 

Fresh leaves 

Fresh twigs and leaf stems 



0.17 
0.12 
0.14 
0.16 
0.16 
0.17 
0.41 
0.76 



0.16 
0.27 
0.51 
0.58 
0.56 
0.51 

0.12 



1.73 
2.56 
3.69 
3.67 
3.78 
4.00 


0.18 



2.06 
2.95 
4.34 
4.41 
4.50 
4.68 
0.41 
1.06 



0.25 
0.52 
0.59 
0.65 
0.80 



0.06 
0.11 
0.11 
0.17 
0.18 



0.34 
1.02 
0.94 
1.16 
1.07 



0.65 
1.65 
1.64 
1.98 
2.05 



The three parent substances producing methyl alcohol occur 
to much less extent in the main root and rootlets than in the 
portion above ground. The pectin methyl alcohol is low in the 
wood in comparison with the lignin methyl alcohol, but in 
the bark the former may attain 30 to 40 per cent of the total. 
The distribution of methyl alcohol in fir wood follows: 

Analysis of Fir Wood 

Per Cent 

Total methyl alcohol 4 . 65 

Methyl alcohol from pectin 0.11 

Methyl alcohol from total lignin 4 . 54 

Methyl alcohol from colored ortholignin 4 . 43 

Methyl alcohol from colored ortholignin acid 

Methyl alcohol from uncolored lignin 0.11 

Pectin content 1.10 

Colored ortholignin 31 .27 

Methyl alcohol content of colored ortholignin 16. 10 

Methyl alcohol content of colored ortholignin acid. ... 

The hydrolysis of the methoxyl groups in lignin takes place 
to only a limited extent in the suphite pulping process and in the 
hydrolysis of wood with dilute acid under pressure as conducted 
in the manufacture of ethyl alcohol. 1 Five grams of Will- 
statter's lignin heated with 30 cubic centimeters of 5.7 per cent 

1 E. Heuser and H. Schmelz, Cellulosechemie, 1 (1920), 49-58. 



114 CHEMISTRY OF CELLULOSE AND WOOD 

hydrochloric acid at 150 to 160° for 2 hours gave, by repeating 
the process, a total of 14.83 per cent methyl alcohol, 2.19 per 
cent acetic acid, and 0.73 per cent acetone. Four hours of 
continuous heating under pressure gave about 13 per cent of 
methyl alcohol, while heating at atmospheric pressure gave only 
2.56 per cent. The yield of acetone is proportional to that of 
methyl alcohol, the maximum obtained being 1 per cent. 

Oxidation of Lignin. — One of the characteristic properties of 
lignin is the ease with which it is decomposed by oxidizing 
agents. With the exception of alkaline fusion, usually the only 
products of oxidation aside from carbon dioxide and water are 
the simple aliphatic acids. 

Willstatter's lignin has a high degree of inflammability. Pine 
lignin, when subjected to a cold stream of oxygen containing 2 
per cent of ozone, burned in 25 seconds. 1 

When oxidized with concentrated nitric acid Willstatter's 
lignin gave 20 per cent of oxalic acid. The yield decreased in 
the presence of catalyzers, such as ammonium vanadate and 
ferrous sulphate. On the other hand, with 25 per cent nitric 
acid the yield was only 4 per cent. This was raised to 17 per 
cent by the addition of ferrous suphate. 2 

Attempts to obtain aromatic acids having a constitutional 
relationship to lignin by the alkaline oxidation of the methyl 
ethers of lignin and ligninsulphonic acid gave negative results. 3 
Methylated ligninsulphonic acid gave 18.57 per cent of oxalic 
acid, and lignin methyl ether 26.40 per cent of oxalic acid and 
100.8 per cent of carbon dioxide. In terms of carbon, only a 
little more than half of the carbon was accounted for in the oxida- 
tion of methyllignin : 44.76 per cent as carbon dioxide and 10.69 
per cent as oxalic acid. Since the volatile acids amounted to 
only 0.9 per cent, it is probable that other volatile products are 
formed. 

Konig and Rump 4 oxidized the lignin and crude fiber of fir 
wood with hydrogen peroxide and ammonia, lime water having 
been added to retain the carbon dioxide formed. With lignin, 

1 E. Erdmann, Z. angew. Chem., 34 (1921), 313. 

2 E. Heuser, H. Roesch, and L. Gunkel, Cellulosechemie, 2 (1921), 13-19. 

3 E. Heuser and S. Samuelsen, Cellulosechemie, 3 (1922), 78-83. 

4 J. Konig and E. Rump, "Chemie — Pflanzen-Zellmembran," p. 61. 



LIGNIN 



115 



formic acid was the principal volatile acid obtained, and with 
the crude fiber, acetic acid. 





Residue, 
per cent 


Formic 

acid, 
per cent 


Acetic 

acid, 

per cent 


Carbon 
dioxide, 
per cent 


Lignin 


10.40 
66.19 


3.46 
4.24 


0.90 
16.07 


8.47 


Crude fiber 


8.43 







Stutzer 1 tried without success to obtain benzene derivatives 
by treating the crude fiber of straw with a nitrating mixture 
(1 volume fuming HN0 3 and 2 volumes concentrated H 2 S0 4 ). 
With this mixture, as well as by oxidation with nitric acid of sp. 
gr. 1.33, small amounts of oxalic, succinic, and suberic acids 
were obtained. Konig 2 worked with barium ligninsulphonate. 
He failed to obtain oxidation products bearing on the constitu- 
tion of lignin by the use of potassium permanganate, electrolytic 
oxidation, hydrogen peroxide, ozone, and nitric acid. The chief 
products of oxidation were water, carbon dioxide, and oxalic 
acid; however, with ozone, formic acid, and with nitric acid, 
succinic acid, were also obtained. Lignin isolated from spruce 
wood with hydrochloric acid (1.21), when suspended in acetic 
acid and oxidized with nitric acid, gave a small amount of a 
pale-yellow, amorphous compound, C19H32O14N. Lignin in 
suspension in acetic acid gave with ozone much greater amounts 
of oxalic acid than moist lignin, owing to moderation of the oxida- 
tion. The formation of an ozonide could not be confirmed by 
ultimate analysis and the use of semicarbazide acetate, the 
latter giving a very small amount of an amorphous precipitate. 3 
This is contrary to the claim of Doree and Cunningham 4 that an 
ozonide is formed from a keto-R-hexene ring. 
CO 



-HC 



H 



CH 2 



H.CK 

I > 
CH.O/ 



COOH 
COOH 



1 A. Stutzer, Ber., 8 (1875), 575-576. 

2 F. Konig, Cellulosechemie, 2 (1921), 105, 117. 

3 F. Konig, Cellulosechemie, 2 (1921), 117. 

4 C. Doree and M. Cunningham, J. Chem. Soc, 103 (1913), 677. 



116 CHEMISTRY OF CELLULOSE AND WOOD 

The formation of formic, acetic, oxalic, malonic, and succinic 
acids by the oxidation of lignin with hydrogen peroxide is con- 
sidered as supporting the theory that lignin contains a coniferyl 
or vanillyl nucleus, succinic acid being also obtained by the oxida- 
tion of vanillin. 1 

Hagglund 2 fused Willstatter's lignin with caustic potash, with 
and without lead peroxide, obtaining a trace of formic acid but 
no acetic or oxalic acid. Catechol could not be detected but 
reactions were obtained for protocatechuic acid. Oxidation with 
alkaline potassium permanganate, and with nitric acid, with and 
without potassium chlorate, gave acetic acid but no oxalic acid. 

Fischer 3 subjected lignin to pressure oxidation in the presence 
of caustic soda, obtaining benzene carboxylic acids. Since 
cellulose under the same conditions gave furan derivatives, he 
concludes that lignin has a benzenoid constitution. Wills- 
tatter's lignin, on oxidation by shaking with air at 200° in the 
presence of 2.5N sodium hydroxide for 40 hours, gave small 
amounts of mellitic acid, benzene pentacarboxylic acid, and oxalic 
acid. 4 Lignin dissolved in dilute sodium carbonate and oxidized 
at 200° with air at 55 atmospheres' pressure gave 8.28 per cent 
of aliphatic acids, and 3.12 per cent of benzene carboxylic acids, 
including benzoic, isophthalic, phthalic, trimellitic, mellitic, 
hemimellitic, prehnitic, benzene tricarboxylic, benzene penta- 
carboxylic, succinic, oxalic, and fumaric. 5 

Since aromatic bodies might be formed during the isolation 
of the lignin with strong hydrochloric acid, pine wood that had 
been extracted with a mixture of alcohol and benzene was oxidized 
at 200° in a current of air at 55 atmospheres' pressure for a long 
time. There was obtained 0.83 per cent of benzene pentacarbox- 
ylic acid, based on the lignin in the wood, in comparison with 1.65 
per cent fromWillstatter's lignin. 6 

1 O. Anderzen and B. Holmberg, Ber., 56 (1923), 2044-2048. 

2 E. Hagglund, Arkiv Kemi Mineral. Geol, 7, 8 (1918), 1-20; Cellulose- 
chemie, 4 (1923), 73. 

3 F. Fischer, N aturwissenschaften, 9 (1921), 958-965. 

4 F. Fischer, H. Schrader, and W. Treibs, Ges. Abhandl. Kenntnis 
Kohle, 5 (1921), 200-210, 221-229. 

5 F. Fischer, H. Schrader, and A. Friedrich, Ges. Abhandl. Kenntnis 
Kohle, 6 (1921), 1-21. 

6 F. Fischer, H. Schrader, and A. Friedrich, Ges. Abhandl. Kenntnis 
Kohle, 6 (1921), 22-26. 



LIGNIN 117 

Lignin when heated with 10JV potassium hydroxide at 300° 
for 3 hours gave phenols, or phenol carboxylic acids, and a small 
quantity of adipic acid. Fusion of lignin with potassium hydrox- 
ide gave 14.9 per cent of substances soluble in ether, among 
which was protocatechuic acid. 1 

Lignin in the presence of 5N caustic soda and oxygen at ordinary 
temperature lost 9.4 per cent of its weight in 46 hours. In 1000 
hours 1 gram of lignin absorbed 82 cubic centimeters of oxygen. 
The products of the oxidation were mainly humic acids, together 
with small amounts of carbon dioxide, formic, acetic, oxalic, 
succinic, isophthalic, and benzene pentacarboxylic acids. 2 

Since cellulose and cane sugar gave small amounts of benzene 
carboxylic acids in addition to furan carboxylic acids, and since 
the yield of benzene carboxylic acids from lignin was only 3 to 
4 per cent, this evidence for a benzenoid structure for lignin is 
not weighty. 

Nitration of Lignin. — Though readily nitrated, no definite 
nitro derivatives have been obtained from lignin. Phelps, 3 
assuming that lignin had aldehydic properties, treated waste 
sulphite liquor with various aromatic amines. Bright-red 
condensation products were obtained with primary amines, and 
brown or black derivatives with other amines. Condensation 
took place in the amino group, since the diazo reaction was unob- 
tainable. Various dyes were obtained on treating the conden- 
sation products with nitric acid, the latter apparently reacting 
with the lignin group. On treating sulphite liquor with nitric 
acid a dye (lignone yellow) was obtained that dyed silk and 
wool orange yellow. The lignin from spent soda liquor after 
treatment with nitric acid dyes wool brown. 4 

Nitro derivatives of ligninsulphonic acid from sulphite liquor 
are prepared by Oman 5 by the use of 50 per cent nitric acid at 30 
to 40°, or with mixed acid at 30°. 

1 F. Fischer and H. Tropsch, Ges. Abhandl. Kenntnis Kohle, 6 (1921), 
271-278; Ber., 66 (1923), 2418. 

2 H. Schrader, Brennstoff-Chem. 3 (1922), 161-167. 

3 E. B. Phelps, U. S. Geol. Survey, Water-supply Paper, 226, Washington 
(1909); cf. W. O. Walker, J. Soc. Chem. Ind., 32 (1913), 390. 

4 M. P. Cram, J. Ind. Eng. Chem., 6 (1914), 896. 

5 E. Oman, E. P. 103651, 103652 (1917). 



118 CHEMISTRY OF CELLULOSE AND WOOD 

Nitrolignin was prepared by treating 150 grams of Wills tatter's 
lignin, with cooling, with 1000 cubic centimeters of 5iV nitric 
acid, then warming until the product became uniformly orange 
colored. The yield was 60 per cent. The yellow powder, after 
solution in alcohol and precipitation with dry hydrogen chloride, 
had the composition C42H37O24N3. It contained three methoxyl 
and six hydroxyl groups. Acetylation yielded a tetracetyl 
derivative. Lignin, owing to its easy nitration, was regarded as 
having a phenolic structure. 1 

Lignin isolated from flax and wood with caustic soda, and 
having the formula C 4 oH 3 o06.(OCH 3 )4.(OH)5.CHO, is readily 
nitrated with nitric acid alone or with an equal part of sulphuric 
acid at —5°. The resulting red powder is a trinitro derivative 
behaving as an aromatic compound. It is easily acetylated. 2 

A nitroso derivative is formed by dissolving 100 grams of sodium 
ligninsulphonate, obtained by the addition of sodium chloride 
to sulphite liquor, in 600 cubic centimeters of water and adding 
60 grams of sodium nitrite in 100 cubic centimeters of water. 
After cooling with ice there is added 60 grams of concentrated 
sulphuric acid in 400 cubic centimeters of water. When the 
mixture has stood for several hours, the nitroso compound is 
precipitated with sodium chloride and washed with a strong 
solution of the same. 3 

When sodium ligninsulphonate is added to diazotized aromatic 
amines, brown to red dyestuffs are obtained. 4 Diazotized 
aniline, benzidine, and /?-napthylamine-disulphonic acid give, 
respectively, brick-red, brown, and red products. 

Distillation of Lignin. — The methyl alcohol and most of the 
acetic acid obtained on distilling wood were considered by Fremy 
and Urbain 5 as derived from the lignin, since lignin (vasculose) 
on distillation gave methyl alcohol. 

Lignin prepared from spruce by the Willstatter method has 
been subjected to destructive distillation by several investigators. 

1 F. Fischer and H. Tropsch, Ges. Abhandl. Kenntnis Kohle, 6 (1921), 
279-288; Chem. Centr., 94, III (1923), 1639. 

2 W. J. Powell and H. Whittaker, J. Chem. Soc., 125 (1924), 357; 127 
(1925), 132. 

3 E. Oman, E. P., 103653, 103654 (1917). 

4 E. Oman, E. P., 103479, 103480; cf. 103822 (1917). 

5 E. Fremy and Urbain, Compt. rend., 94 (1882), 110. 



LIGNIN 



119 



Hagglund, 1 
per cent 



Heuser, 2 
per cent 



C0 2 

CO 

CH 4 

Cnll2re 

H 2 

2 

N (and undetermined) 

Tar 

Methyl alcohol 

Acetone 

Acetic acid 

Charcoal 



23 

17 

12.5 
1.4 

13.6 
2.5 

30.0 
9.6 
0.7 
0.1 
0.6 

45 



9.60 
50.90 
37.50 

2.00 



13.00 
0.90 
0.19 
1.09 

50.64 



\ By volume 



By weight on the 
dry, ash-free lig- 



1 E. Hagglund, Arkiv Kemi, Mineral. Geol., 7, 2 (1918), 19. 

2 E. Heuser and C. Skioldebrand, Z. angew. Chem., 32 (1919), 41. 



The discrepancy in the results may be due to different experi- 
mental conditions, and the probability that Hagglund's lignin, 
owing to its high furfural content (3.69 per cent), was not free 
from carbohydrates. In the light of Heuser' s results in partic- 
ular, lignin, in comparison with spruce wood, gives larger amounts 
of methane, carbon monoxide, charcoal, and tar, and less carbon 
dioxide. During distillation the methoxyl groups appear to be 
largely decomposed into gases, since the yield of methyl alcohol 
from the lignin is only about one-third of that of the original 
wood. The low yield of acetic acid is due in part to saponifica- 
tion of the acetyl groups during the isolation of the lignin. 
The lignin, the reaction being exothermic, began to char at 270° 
and decomposed to the greatest extent at 400 to 450°, though 
gases were formed up to 627°. 

Lignin gives about one-half as much tar and nearly twice as 
much coke on distillation as does cellulose. 1 The lignin tar 
contained 50 per cent of acid and 13 per cent of neutral bodies, 
in comparison with 17 and 31 per cent, respectively, for the 
cellulose tar. Erdmann, 2 on distilling lignin from Scotch pine, 

1 F. Fischer and H. Schrader, Ges. Abhandl. Kenntnis Kohle, 6 (1920), 
106-116. 

2 E. Erdmann, Z. angew. Chem., 34 (1921), 309. 



120 



CHEMISTRY OF CELLULOSE AND WOOD 



obtained 18.1 per cent of tar containing 37 per cent of creosote. 
The total yield of creosote based on the lignin was 7.6 per cent. 
Both cellulose and lignin gave phenols on distillation, the latter, 
however, seven to eight times as much as the former. In the 
distillation water from lignin, a strong reaction for catechol 
was obtained, while in that from cellulose no catechol was 
detected. 

Pictet and Gaulis 1 distilled lignin from fir wood at 350 to 390° 
under 25-millimeters pressure, obtaining 15 per cent of tar. On 
removing the acids and phenols, 2 per cent of the lignin was 
obtained as a brown oil. The latter was separated into saturated 
and unsaturated hydrocarbons by means of liquid sulphur dioxide 
and fractioned. The third, sixth, and eighth fractions appeared 
to be identical with corresponding coal-tar fractions. The com- 
pound Ci 3 Hi6, which is similar to the hexahydrofluorene from 
coal tar, gave a tetrabromo-derivative, CisH^Br*; m.p. 193°. 



Fraction No. 


Boiling 
point 


Formula 


Specific 
gravity 


Refractive 
index 


1 unsaturated 


235-240° 
260-270 
270-280 
315-320 
Above 320 
200-210 
230-240 
250-260 


C13H26 
C14H26 

C16H30 
C24H44 

C 30X160 

C n H 16 

C12H16 
C13H16 


0.8091 
0.8138 
0.8218 
0.8579 

0.8964 
0.9172 
0.9372 


1 . 4468 


2 


1 . 4532 


3 


1 . 4541 


4 




5 




6 saturated 


1.5119 


7 


1 . 5226 


8 


1 . 5422 







The fifth fraction was identified as melene, 2 C 3 oH 6 o; m.p. 62 to 
63°. Eugenol, m.p. 68°, was identified in the phenol fraction. 
The assumption that all the above fractions belong to the hydro- 
aromatic series and that lignin contains a hydroaromatic ring 
seems hardly justified in view of the fact that the hydrocarbons 
represent but 2 per cent of the lignin distilled. 



1 A. Pictet and M. Gaulis, Helvetica Chim. Acta, 6 (1923), 627-640. 

2 Melene, though formerly considered to be an alkylene hydrocarbon, is 
held by Pictet (Ber., 48 (1915), 933) to be a cyclic, saturated compound. 



LIGNIN 121 

There is no close relationship between lignin and cellulose, 
since on distillation under reduced pressure no laevoglucosan is 
obtained. 1 

Lignin on distillation with zinc dust gives about 17 per cent 
of a viscous oil distilling between 66 and 280° at 2-millimeters 
pressure. All of the fractions contain methoxyl. 2 

1 F. Fischer and H. Tropsch, Naturprodukte, 24 (1923), 8; Ber., 56 (1923), 
2418. 

2 P. Karrer and B. Bodding-Wiger, Helvetica Chim Acta., 6 (1923), 
817-822. 



CHAPTER IV 
COLOR REACTIONS OF WOOD 

Wood gives a large number of color reactions with various 
inorganic and organic compounds. A few of the reactions are 
specific for lignin, while the majority are due to the presence of 
traces of aldehydes, adsorption, and other phenomena. 

Color Reactions of Wood with Inorganic Compounds. — Potas- 
sium permanganate is rapidly reduced by lignocelluloses with 
the deposition of brown hydrated manganese dioxide. The 
extent to which fibers turn brown has been employed as a criterion 
of their technical value. 1 The greenish color obtained with 
ferric chloride is evidently due to traces of tannin. A 15 to 40 
per cent solution of cobalt sulphocyanate colors lignified mem- 
branes blue, the phenomenon being apparently due entirely to 
adsorption. Casparis 2 considers this reaction more sensitive 
than the Maule, or the phloroglucinol reaction. 

Zinc chloriodine solution, which gives a yellow color with 
lignified tissues, is much used for staining wood pulp. Schorger 3 
has found, however, that moist gelatinized wood is instantly 
stained a deep blue black. Sections of wood after boiling in 
water are also stained blue. 4 

In 1893, Cross and Be van 5 found that lignocelluloses react 
with a ferricferricyanide solution to form Prussian blue, which 
is intimately attached to the fiber. The reaction is considered 
to take place specifically between the fiber and the ferricferri- 
cyanide solution, since lignocellulose has little action on ferric 
chloride or potassium ferricyanide solution taken separately. 

1 A. Aisslinger, "Beitrage zur Kenntnis wenig bekannter Pflanzen- 
fasern," Diss. Zurich (1907), p. 13. 

2 P. Casparis, C. A., 15 (1921), 1333. 

3 A. W. Schorger, J. Ind. Eng. Chem., 15 (1923), 814. 

4 M. C. Potter, Ann. Botany, 18 (1904), 121. 

5 C. F. Cross and E. J. Bevan, J. Soc. Chem. Ind., 12 (1893), 104-106; 
"Cellulose," (1916), pp. 124-131. 

122 



COLOR REACTIONS OF WOOD 123 

Aldehyde groups in the lignocellulose are believed to be the cause 
of the reaction, though the presence of oxidizing agents does not 
prevent the formation of the blue precipitate. The reaction is 
not specific for lignin, since it is given by raw cotton. Haller 1 
found that the pigment was deposited only in the cuticular 
layers and attributes the reaction to phenolic hydroxyl groups. 
Hydrocellulose and oxycellulose were without action. According 
to Crocker, 2 the reaction is obtained with any amine or phenol 
and with inorganic reducing agents. No reaction was obtained 
with benzaldehyde and formaldehyde. 

The soft tissues of sections of wood treated momentarily with 
ferric sulphate, then washed with water, dilute acetic acid, and 
water, when immersed in potassium ferrocyanide are colored 
blue. The color is accentuated by the addition of a drop of 
hydrochloric acid. Devaux 3 considered this to be a specific 
reaction for pectin, an iron pectate being first formed. Lignified 
tissues are only weakly colored until after they have been sub- 
jected to the action of potassium hypochlorite, when they are 
strongly colored, due to the formation of oxycellulose. 4 Mangin 5 
found that the middle lamella of wood cells, especially after 
treatment with alcohol-hydrochloric acid, was stained by ruthe- 
nium red. He also believed the latter to be a specific reagent for 
pectin, and that the middle lamella consists mainly of this 
substance. 

Wood on treatment with chlorine becomes yellow, due to a 
combination of the chlorine with the lignin ; if a dilute solution of 
sodium sulphite is applied, as in the Cross and Bevan method for 
determining cellulose, the liquid is colored pink to purple red. 6 
The angiosperms give a more intense color than the gymno- 
sperms. Payen 7 as long ago as 1849 observed that if various 
fibers were immersed for a short time in a saturated chlorine 



1 R. Haller, Fdrber-Ztg., 26 (1915), 157, 173; 30 (1919), 29, 43. 

2 E. C. Crocker, J. Ind. Eng. Chem., 13 (1921), 625. 

3 H. Devaux, Actes Soc. Linn. Bordeaux, 6 (1901), XXXIII-XXXV. 

4 H. Devaux, Actes Soc. Linn. Bordeaux, 6 (1901), LVIII-LIX. 

5 L. Mangin, Compt. rend., 116 (1893), 653. 

6 E. J. Bevan and C. F. Cross, Chem. News, 42 (1880), 80; C. S. Webster, 
J. Chem. Soc, 43 (1883), 25. 

7 A. Payen, Compt. rend., 29 (1849), 493. 



124 CHEMISTRY OF CELLULOSE AND WOOD 

solution and then treated with ammonia, colorations were 
obtained; hemp, e.g., gave a bright red. 

A reaction having the same fundamental principle of oxidation 
was described by Maule. 1 Lignified material when left in a 1 
per cent solution of potassium permanganate for 5 minutes, 
washed, treated with dilute hydrochloric acid, washed, and then 
subjected to the action of ammonia, acquired a deep-red color. 
The writer 2 found that the angiosperms gave the typical red 
color, while the gymnosperms gave only an indefinite brownish 
gray. The Maule reaction was applied by Sharma 3 to 40 species 
of angiosperms and 27 species of gymnosperms; with the latter 
a red color was in no case obtained, the reaction thus serving to 
distinguish between the woods of the two botanical subdivisions. 
An important earlier paper had been overlooked. De Lamar- 
liere 4 studied the Maule reaction in 1903 and found that a red 
color was never obtained with gymnosperms. He attributed 
the reaction to oxidized lignin, having observed that the intensity 
of color obtained with the Maule reaction increased, while that 
with phloroglucinol decreased with the rigor of the oxidation; 
hence the two reactions were not identical. Fuming nitric acid 
and potassium hypochlorite did not give as good results as potas- 
sium permanganate. A 5 per cent solution of chromic acid acting 
for 5 minutes was just as effective, but Hoffmeister's reagent 
(KC10 3 + HC1) when rich in free chlorine was the best. He 
found that in the Maule reaction the hydrochloric acid could 
be replaced by sulphuric or phosphoric acid, and the ammonia by 
potassium carbonate and other alkalis. This shows the close 
relationship between the Maule reaction and the chlorine-sodium 
sulphite reaction, sodium sulphite being weakly alkaline. 
Recently, Casparis 5 and Crocker 6 reached a similar conclusion. 

A tedious lignin reaction has been described by Combes. 7 
The material is treated with a hypochlorite solution for 15 

1 C. Maule, Funf stuck' s Reitrage z. wiss. Botan., 4 (1900), 166. 

2 A. W, Schorger, J. Ind. Eng. Chem., 9 (1917), 561. 

3 F. D. Sharma, /. Forestry, 2 (1922), 476-478. 

4 L. G. de Lamarliere, Rev. gen. botan., 15 (1903), 149-159. 

5 P. Casparis, I.e. 

6 E. C. Crocker, I.e. 

7 R. Combes, Bull. Sci. pharmacol, 13 (1906), 293-296. 



COLOR REACTIONS OF WOOD 125 

minutes, washed, heated on the water bath for 15 minutes with 
1 gram of zinc oxide (or lead nitrate or acetate), suspended in 
30 cubic centimeters of water, washed, treated with hydrogen 
sulphide for 10 to 15 minutes, and finally with a few drops of 
sulphuric acid. There results a fine red color changing to orange- 
red, and finally to brown in 8 to 10 hours. 

Some of the inorganic acids give green colorations. Stolba 1 
noticed that if a softwood was moistened with dilute sulphuric 
acid, it became after some time a beautiful moss green, the color 
being destroyed by water or alkalis. Concentrated hydro- 
chloric acid also colors wood green. 2 The intense green color 
given wood by fuming hydrobromic acid is more permanent 
than that given by fuming hydrochloric acid and compares with 
the phloroglucinol reaction in delicacy. 3 Linde 4 found that 
sections of dry coniferous wood when placed in sulphuric acid 
(65 per cent) acquire a yellow color changing to green; if then 
placed in water the color changes to blue, finally becoming color- 
less. If the wood is first treated with a 5 per cent solution of 
caustic soda, the acid produces only a grayish-brown or gray 
color. The coloration produced by the acid varies with its 
strength and the species of wood, some woods giving only a 
yellow color. Old bast cells of a stem or root of conifers and 
cycads acquire a fine rose color when moistened with a drop of 
concentrated hydrochloric acid. 5 According to Guillard, 6 the 
coloration is obtained not only with bast and cork cells but 
with the structural elements of wood as well. The reaction 
has been attributed to the presence of coniferin. 

Vanadium pentoxide dissolved in phosphoric acid stains wood 
a yellowish brown and gives a reddish-brown precipitate with 
vanillin. 7 Removal of the wood gum increases the intensity 



1 F. Stolba, J. prakt. Chem., 90 (1864), 466. 

2 H. Warnecke, Pharm. Ztg., 33 (1888), 574. 

3 P. Friedlander, Papier-Zeit., 25 (1900), 1877; H. Seidel, Mitt. k.k. 
techn. Gewerbe-Mus., 10 (1900), 37-38. 

4 O. Linde, Arch. Pharm., 244 (1906), 57-62. 

5 P. van Tieghem, Compt. rend., 56 (1863), 963-965. 

6 A. Guillard, Compt. rend., 56 (1863), 1126-1128; F. von Hohnel, 
Akad. Wiss. Wien, 76, I (1877), 676; see Ibid., p. 693 for older literature. 

7 J. Gruss, Ber. botan. Ges., 38 (1920), 361-368. 



126 CHEMISTRY OF CELLULOSE AND WOOD 

of the stain. Addition of vanadium pentoxide to the wood does 
not prevent staining with phloroglucinol. 

Color Reactions of Wood with Organic Compounds. — A 

large number of color reactions with various organic bodies have 
been recorded; compounds containing amino and phenolic 
groups are the most important. A high intensity of color is 
usually obtained at once with the amines, while with the phenols 
light is of great assistance in bringing out the color. 

In 1834, Runge 1 found that spruce wood, or a substance 
extractable therefrom with water and alcohol, gave a blue color 
when treated with a mixture of phenol and hydrochloric acid. 
It was later found that the reaction was not limited to coniferous 
woods, and that any of the common inorganic acids, but not 
organic acids, could be used in place of hydrochloric acid. 2 
While concentrated hydrochloric acid alone may give a green 
color, this is not true of all woods; in the majority of cases the 
coloration is much more intense in the presence of phenol. 

Von Hohnel, 3 in 1877, found that if an aqueous or alcoholic 
extract of cherry wood was applied to lignified tissues, then 
moistened with hydrochloric acid, a red to violet color was 
produced. The body causing the color was called xylophilin. 
Wiesner 4 considered xylophilin to be a mixture of phloroglucinol 
and a little catechol, and attributed the reaction mainly to 
phloroglucinol. He found that a 0.5 per cent solution of phloro- 
glucinol, followed by hydrochloric acid, colored red even a trace 
of woody fiber. The color varies from cherry red to reddish 
violet. 

The colorations given by various phenols and phenol ethers 
in the presence of hydrochloric acid are: 

1 F. F. Runge, J. prakt. Chem., 1 (1834), 24. 

2 R. von Wagner, J. prakt. Chem., 51 (1850), 95. 

3 F. von Hohnel, Akad. Wiss. Wien, 76, I (1877), 663-716. 

4 J. Wiesner, Akad. Wiss. Wien, 77, I (1878), 60-66; Dinglers polytech. 
J., 227 (1878), 397. 



COLOR REACTIONS OF WOOD 



127 



Color Reactions of Wood with Phenols 



Reagent 


Color s Author 


Pyrogallol 


Greenish blue 
Bluish violet 
Bluish violet 
Green 

Bluish violet 
Bluish green 
Bluish green 
Yellowish green 
Greenish 
Greenish yellow 
Greenish yellow 
Green 


Wiesner 1 


Catechol 


Wiesner 1 


Resorcinol 


Wiesner 1 


Salicylic acid 


Warnecke 2 


Orcinol 


Von Lippmann 3 


a- and /3-Napthol 


Schaeffer 4 


Thymol 


Molisch 5 


Guaiacol 


Czapek 6 


Cresol 


Czapek 6 


Anisol 


Czapek 6 


Anethol 


Czapek 6 


Hydroquinone 


Warnecke 2 







1 J. Wiesner, cited by V. Grafe, Monatsh., 25 (1904), 990; cf. F. von Hohnel, Akad. 
Wiss. Wien, 76, I (1877), 716. 

2 H. Warnecke, Pharm. Ztg., 33 (1888), 574. 

3 E. O. von Lippmann, cited by Grafe, I.e. 

4 L. Schaeffer, Ber., 2 (1869), 91; A. Ihl, Chem. Ztg., 9 (1885), 226. 
6 H. Molisch, Ber. botan. Ges., 4 (1886), 303-305. 
6 F. Czapek, Z. physiol. Chem., 27 (1899), 147. 

The addition of potassium chlorate to the hydrochloric acid 
used with the phenol has been observed to hasten the color 
reaction. The Tommasis 1 found that by means of a pine 
splinter moistened in a solution consisting of 50 cubic centimeters 
of hydrochloric acid, 50 cubic centimeters of water, and 0.2 
gram of potassium chlorate, it was possible to detect 1 part of 
phenol in 6000 parts of water. 

Wood treated with pyrogallol in dilute glycerine solution and 
sulphuric acid, or zinc chloride, gives a violet color stable to 
water and hot alcohol. 2 Pure phenol ethers, such as anethol 
and anisol, do not color wood at all; 3 a coloration with these com- 
pounds is without doubt due to the presence of traces of phenol. 

The inorganic salts of organic bases give yellow, green, and 
red stains. Runge 4 was the first to notice that aniline sulphate 

1 T. and D. Tommasi, Ber., 14 (1881), 1834-1836; F. von Hohnel and 
H. Molisch, I.e. 

2 C. Reichl, Rep. anal. Chem., 3 (1883), 125. 

3 E. C. Crocker, J. Ind. Eng. Chem., 13 (1921), 625; O. Adler, Biochem. 
Z., 128 (1922), 32. 

4 F. F. Runge, Pogg. Ann., 31 (1834), 65. 



128 CHEMISTRY OF CELLULOSE AND WOOD 

colored wood yellow. This reagent was subsequently proposed 
by Schapringer 1 for the detection of wood pulp in paper and by 
Wiesner 2 for determining lignification in plant anatomy. 

The colorations obtained with various amino derivatives are 
listed on the following page. 

According to Crocker, no color is obtained with formamide, 
acetamide, urea, or acetanilid. 

Wood suspended in dilute soda solution is colored a brick red 
by the addition of a very small amount of diazobenzene chloride. 
Some of the hardwoods do not react. 3 

The reaction between dimethyl-p-phenylenediamine and wood 
is not a definite one. 4 The intensities of the colorations and the 
amounts of the base fixed, when wood flour was treated with 
increasing proportions of this base in equal volumes of water, 
were proportional to the quantities of the base taken, an equilib- 
rium being reached. This reagent should be used with caution 
for the detection of lignified fibers, particularly in paper, since 
it reacts very readily with oxidizing agents, such as ozone, 
chlorine, and peroxides, a red color being produced. 5 The red 
color produced with wood is discharged by a drop of ammonia 
and restored by glacial acetic acid, while if the color has been 
produced by chlorine it remains permanently discharged by this 
treatment. 

Paranitraniline, 0.2 gram in 100 grams of 20 per cent sulphuric 
acid, colors lignin a reddish orange. 6 Wheeler 7 states that a 2 
per cent solution of the nitranilines in 12 per cent hydrochloric 
acid gives a blood-red color, the p-compound being the most 
sensitive. According to Grandmougin, 8 p-amido-diphenylamine 
gives a Bordeaux red, resistant to acids, and much more sensitive 
than p-nitraniline. If the amino groups are diazotized or meth- 
ylated, the coloring power is greatly reduced or destroyed. 9 It 

1 S. Schapringer, Dinglers polytech. J., 176 (1865), 166. 

2 F. Wiesner, Karsten's hot. Untersuchungen, 1 (1866), 120. 

3 E. Ungar, Diss. Budapest (1916), p. 49. 

4 C. F. Cross, E. J. Bevan, and J. F. Briggs, Ber., 40 (1907), 3119. 

5 C. Wurster, Papier-Zeit., 28 (1903), 1608-1609. 

6 A. Berge, Bull. Soc. Chim. Belg., 20 (1906), 158-159. 

7 A. S. Wheeler, Ber., 40 (1907), 1888-1890. 

8 E. Grandmougin, Ber., 40 (1907), 2453. 

9 E. Grandmougin, Z. Farben-u. Textilchemie, 5 (1906), 321-323. 



COLOR REACTIONS OF WOOD 



129 



Color Reactions of Wood with Organic Bases 



Reagent 



Color 



Author 



p-Toluidine 

o-Toluidine 

Benzidine 

p-Aminobenzylidenacetophenone 

o-Bromphenetidine 

o-Bromanisidine 

o- and p-Amidophenol 

p-Anisidine 

p-Phenetidine 

m- and p-Phenylaminophenol . . . 

Aminoanthracene 

Morphine 

Acetanilid 

Phenacetin 

Xylidine 

a- and /S-Napthylamine 

m-Phenylenediamine 

p-Phenylenediamine 

Dimethyl-p-phenylenediamine. . . 

Toluylenediamine 

Diphenylamine 

Urea 

Antipyrine 

Phenylhydrazine 

p-Amidodiphenylamine 

Hydrazine 

Sulphanilic acid 

o-, m-, and p-Chloraniline 

o- and p-Aminophenol 

Orthof orm 

Aminosalicylic acid 

o- and p-Aminobenzoic acid 

o- and w-Nitraniline 

p-Nitraniline 

Triaminophenol 



Yellow 


Singer 1 


Yellow 


Covelli 2 


Yellow to red 


Dukelskys 


Brownish red 


Rupe 4 


Yellow 


Piuttis 


Yellow 


Piuttis 


Yellow 


Piuttis 


Yellow 


Piuttis 


Yellow 


Piuttis 


Yellow 


Piuttis 


Blood red 


Kielmeyer 6 


Green 


Warnecke 7 


Green 


Warnecke 7 


Green 


Warnecke 7 


Yellow 


Nickel 8 


Yellow 


Nickel 8 


Yellow 


Molisch* 


Orange red 


Covelli 10 


Red 


Wurster 11 


Dark orange 


Hegleri2 


Golden orange 


Ellramis 


Yellow 


Ihl 14 


Red 


Ihl 14 


Yellow, then green 


Nickel 15 


Bordeaux red 


Grandmougin 


Yellow to orange 


Nickel 17 


Orange yellow 


Covelli 18 


Orange yellow 


Covelli 18 


Orange yellow 


Covelliis 


Orange yellow 


Covelli m 


Orange yellow 


Covelli 18 


Orange yellow 


Covelli 18 


Orange red 


Covelli 18 


Reddish orange 


Covelli 18 


Orange red 


Covelli 18 



1 M. Singer, Monatsh., 3 (1882), 408. 

2 E. Covelli, Chem. Ztg., 25 (1901), 684. 

3 S. Dukelsky, Papier-Fabr., 10 (1912), 6; H. Schneider, Z. wiss. Mikrosk., 31 (1914), 68. 

4 H. Rupe and A. Porai-Koschitz, Z. Farben-Ind., 5 (1906), 317. 
6 A. Piutti, Gazz. Chim. Ital., 28, II (1898), 168-170. 

6 E. Kielmeyer, Dinglers polytech. J., 227 (1878), 584-558. 

' H. Warnecke, Pharm. Ztg., 33 (1888), 573. 

8 E. Nickel, "Die Farbenreaktionen der Kohlenstoffverbindungen" (1890), p. 51. 

* H. Molisch, Sitzb. zool. botan. Ges. Wien, 37 (1887), 30. 

»° E. Covelli, Chem. Ztg., 25 (1901), 684; H. Blau, Pharm. Post, 38 (1905), 753. 
» C. Wurster, Ber., 20 (1887), 808-810. 
i 2 R. Hegler, Flora, 73 (1890), 35. 
" W. Ellram, Chem. Centr., II (1896), 99. 
14 A. Ihl, Chem. Ztg., 13 (1889), 831. 
is E. Nickel, Chem. Ztg., 17 (1893), 1209; E. 
Biochem. Ztg., 128 (1922), 32. 

" E. Grandmougin, Ber., 40 (1907), 2453. 
» E. Nickel, Chem. Ztg., 17 (1893), 1243. 
18 E. Covelli, I.e. 



Senft, Monatsh., 25 (1904), 419; O. Adler, 



130 CHEMISTRY OF CELLULOSE AND WOOD 

had been found by Covelli 1 that if one of the hydrogen atoms in 
the amino group is replaced by alkyl or phenyl groups, the amine 
retains the power of coloring wood, but to a less degree; however, 
it vanishes on introducing an acid radical. Wood treated with 
phenylhydrazine hydrochloride becomes yellow, then red, and 
finally green. If the wood has been given a preliminary treatment 
with chlorine, it is no longer colored by amines. The change in 
color of the amine condensation products may be due to photo- 
trophy. Among the anils studied by Senier and Shepheard, 2 
salicylidene-m-toluidine changed from pale yellow to deep 
orange on exposure to sunlight, the change being reversible. 
The green color given by phenylhydrazine is considered by 
Jentsch 3 as more definite than the phloroglucinol red for deter- 
mining the extent to which wood pulp has been freed from lignin. 
Some of the aromatic bases with nitrogen in the ring give deep 
colorations with lignin. Runge 4 found that hydrochloric acid 
and pyrrol turned wood a dark purplish red. The reaction 
has been rediscovered several times. 5 It is a sensitive reagent 
for lignin, but has the disadvantage of decomposing readily. 6 
Streeb 7 used a-methylindol and obtained a dark red with lignified 
material. According to Renker, 8 oxy cellulose is also colored 
red by this reagent. A saturated solution of thalline sulphate 
(p-methoxytetrahydroquinoline sulphate, (C 6 Hi 3 NO)2.HS04 + 
2H 2 0) colors lignin a dark orange yellow, while cellulose and 
cork are not colored. 9 The reagent is very sensitive and gives 
a coloration even in a 0.01 per cent solution. The red color 
given by pyrrol, indol, and carbazol is destroyed by ammonia. 10 

1 E. Covelli, I.e. 

2 A. Senier and F. G. Shepheard, J. Chem. Soc., 96 (1909), 441-445; 
1943-1955. 

3 S. Jentsch, Z. angew. Chem., 31 (1918), 72. 

4 F. F. Runge, J. prakt. Chem., 1 (1834), 24. 

5 N. Lubavin, Ber., 2 (1869), 100; A. Ihl, Chem. Ztg., 14 (1890), 304, 
1571; M. Niggl, Chem. Ztg., 15 (1891), 298. 

6 M. Singer, Monatsh., 3 (1882), 398. 

7 E. Streeb, "Ueber Derivate des Lignin's," Diss. Gottingen (1892). 

8 M. Renker, " Bestimmungsmethoden der Cellulose," Berlin (1910), 
p. 26. 

9 R. Hegler, Botan. Centralbl, 38 (1889), 616-618; Flora, 73 (1890), 
31-61. 

10 E. Covelli, Chem. Ztg., 25 (1901), 684. 



COLOR REACTIONS OF WOOD 



131 



Reagent 


Color 


Author 


Bone oil 


Red 

Red 

Pink 

Yellow 

Yellow 

Red 

Violet 

Reddish violet 

Cherry red 


Ihl 1 




Ihl 2 


Pyridine 


Ihl 3 


Picoline 


Ihl 3 


Lutidine 


Ihl 3 




Ihl 4 


Skatol 


Mattirolo 5 


Carbazol 


Mattirolo 5 


Indol 


Niggl 6 







i A. Ihl, Chem. Ztg., 14 (1890), 34. 

2 A. Ihl, Chem. Ztg., 14 (1890), 67. 

3 A. Ihl, Chem. Ztg., 14 (1890), 304. 

4 A. Ihl, Chem. Ztg., 14 (1890), 1571. 



5 D. O. Mattirolo, Z. wiss. Mikrosk., 2 (1885), 354-355. 

6 M. Niggl, Flora, 64 (1881), 545-547; cf. A. Baeyer, Ann., 140 (1866), 296. 

Crocker 1 states that carbazol and pyridine give no coloration. 

Color reactions are also obtained by the use of aliphatic 
alcohols and sulphuric acid. When equal volumes of amyl 
alcohol, free from furfural, and sulphuric acid are heated at 
about 90° until evolution of gas begins, a reddish-yellow solution 
is obtained which stains wood fibers red to deep blue. 2 The 
color is attributed to furfural formed from xylose. Linde 3 
states that the reaction is misleading since the same coloration 
is produced with asbestos. All the alkyl sulphuric acids and 
aryl sulphonic acids give a red or blue color with lignin; naptha- 
lene sulphonic acid gives a blue and anthracene sulphonic acid 
a deep-red color. 4 Benzene heated with sulphuric acid until 
sulphur dioxide is evolved gives an intense blue with wood. A 
reddish-violet color is obtained with isobutyl or hexyl alcohol, 
or isobutyl aldehyde and sulphuric acid in the presence of lignin ; 
the reaction is not obtained with hydrochloric acid. 5 



1 E. Crocker, I.e. 

2 A. Kaiser, Chem. Ztg., 26 (1902), 335. 

3 O. Linde, Arch. Pharm., 244 (1906), 62. 

4 J. Hertkorn, Chem. Ztg., 26 (1902), 632. 

5 V. Grafe, Oesterr. botan. Zeits., 55 (1905), 174-176. 



132 CHEMISTRY OF CELLULOSE AND WOOD 

Thiophene 1 stains wood green and methylheptenone 2 red in 
the presence of hydrochloric acid. Reichl 3 found that egg 
albumin dissolved in dilute sulphuric acid to which a little 
ferrous sulphate was added, when heated, colored spruce shavings 
green, while the liquid became violet. Wood is colored violet 
when dipped into oil of myrrh, 4 partially dried, and then 
immersed in dilute hydrochloric acid; sesame oil 5 gives an 
intense green. 

The various amines and phenols were found by Barrett 6 
to be unsuited for the detection of lignified impurities in 
cotton to be used for nitration, preference being given to 
staining with malachite green. A solution is prepared by 
dissolving 0.1 gram of the dye in 100 cubic centimeters of hot 
water, diluting to 500 cubic centimeters, adding 50 cubic centi- 
meters of 40 per cent formaldehyde solution, 1 gram of sodium 
bisulphate, and making up to 1 liter. A second solution con- 
sists of 20 grams of bleaching powder in 1 liter of water. The 
sample (3 grams) is heated with 300 cubic centimeters of the 
dye solution on the water bath for 10 minutes and then 25 cubic 
centimeters of the clear hypochlorite solution added with rapid 
stirring. After 5 minutes the sample is removed and washed. 
The ligneous impurities are stained, while the color on the 
cotton is completely discharged. Bugnon 7 treated lignified 
sections with sodium hypochlorite, washed, and then immersed 
them in a solution of light-green F. S. (sodium diethyl-dibenzyl- 
diamino-triphenylcarbinol-sulphonate) in 5 per cent hydrochloric 
acid. A permanent green color was obtained. Lignified tissues 
are stained a deep red by an aqueous solution of fuchsin; in 
distinction from non-lignified tissues, maceration in glycerine 
does not remove the color. 8 

The nature of the substances in lignified tissues producing 
the various color reactions is obscure. The phenol reaction was 

1 A. Ihl, Chem. Ztg., 14 (1890), 1707. 

2 E. and H. Erdmann, Ber., 32 (1899), 1218. 

3 C. Reichl, Monatsh., 11 (1890), 159. 

4 O. Linde, Arch. Pharm., 244 (1906), 62. 

5 H. Kreis, Chem. Ztg., 28 (1904), 956-957. 

6 F. L. Barrett, J. Soc. Chem. Ind., 39 (1920), 81-82T. 

7 P. Bugnon, Compt. rend., 168 (1919), 62-64. 

8 G. Dragendorff, "Plant Analysis," London (1884), p. 95. 



COLOR REACTIONS OF WOOD 133 

first attributed to the glucoside coniferin. In 1861, Hartig 1 
discovered this compound in the cambium sap of Larix europea 
and called it "laricin." Kubel 2 gave coniferin its present name, 
having found it to be present in the cambium sap of all of the 
conifers examined. Coniferin, Ci 6 H 2 20 8 + 2H 2 0, crystallizes 
in satiny needles that lose their water of crystallization at 100° 
and melt at 185°. It is very slightly soluble in cold water, 
readily so in hot, and insoluble in ether. On hydrolysis with 
emulsin it gives d-glucose and coniferyl alcohol; dilute acids on 
heating produce resinification; oxidation with chromic acid 
mixture gives vanillin. Coniferin and vanillin give a red color 
with phloroglucinol-hydrochloric acid. 

Tiemann and Harmann 3 found that coniferin gives with con- 
centrated sulphuric acid a dark- violet color, changing to blue on 
the addition of water. Concentrated hydrochloric acid in the 
cold gives no coloration, but on heating a blue precipitate is 
produced. The young wood of conifers, on moistening with 
concentrated sulphuric acid, is colored violet. Since coniferin 
and phenol with concentrated hydrochloric acid, in the sunlight, 
were almost instantly colored a deep blue, the bluish color 
obtained with phenol-hydrochloric acid and pine wood was 
attributed to traces of coniferin in the latter. Soon afterwards 
Tangl 4 applied the phenol reaction to angiosperms and, having 
obtained the same bluish green as with the conifers, decided that 
coniferin must be widely distributed in nature. 

The characteristic "wood substance" was considered by 
Singer 5 to contain vanillin and coniferin, since the hot-water 
extract had the odor of the former and gave some of the color 
reactions of the latter. Lignin was looked upon as a mixture of 
several substances and not as a definite compound. He found 
that old, fine spruce sawdust had to be boiled with water for 18 
days before it ceased to give a blue color with phenol-hydrochloric 
acid. Based on the slender evidence of odor or color reactions, 

1 T. Hartig, Jahrb. f. Forster, 1 (1861), 263. 

2 W. Kubel, J. prakt. Chem., 97 (1866), 243-246. 

3 F. Tiemann and W. Harmann, Ber., 7 (1874), 608-623. 

4 E. Tangl, Flora, 57 (1874), 239. 

5 M. Singer, Monatsh., 3 (1882), 395-410. 



134 CHEMISTRY OF CELLULOSE AND WOOD 

various observers have reported the presence of vanillin in wood, l 
and in sulphite 2 and soda 3 liquor from the manufacture of wood 
pulp. Fresh linden bark 4 collected in summer contains detect- 
able amounts of vanillin, though bark collected from the same 
tree during the winter months was free from it. The presence 
of vanillin in decayed oak wood 5 has been reported on the basis 
of the blue color obtained by the use of the phosphotungstic- 
phosphomolybdic acid solution of Folin and Denis. 6 The most 
trustworthy occurrence of vanillin is that mentioned by von 
Lippmann, 7 who found a sufficient deposit on a board to deter- 
mine that it gave a blue color with ferric chloride and had a m.p. 
of 81°. It is possible that under certain pathological conditions 
vanillin might be formed from coniferin. 

Since diphenylamine gave an intense greenish yellow with 
vanillin and a golden yellow with coniferin, Ellram 8 concluded 
that coniferin predominated in the membranes of the youngest 
cells, and vanillin in the oldest lignified tissues. Hegler 9 con- 
sidered coniferin to be always present in lignified membranes; 
according to Richter, 10 the best reaction for wood substance is 
the test with phloroglucinol-hydrochloric acid and its cause is 
vanillin. Von Udranzky 11 found that the woods of Pinus sylves- 
tris, Abies excelsa, and A. pectinata were colored yellow or greenish 
with concentrated hydrochloric acid in the cold, and reddish 
yellow, red, or bluish green on heating, the color being especially 
intensive at the border of the annual rings. Concentrated 

1 W. Hoffmeister, Landw. Jahrb., 17 (1888), 260; V. Grafe, Monatsh., 
25 (1904), 1014. 

2 J. B. Lindsey and B. Tollens, Ann., 267 (1892), 341; H. Seidel, Z. 
angew. Chem., 11 (1898), 876; E. Pollacsek, Oesterr. Privilegium, 1524 
(1898); cf. M. Muller, "Literatur der Sulfit-Ablauge," Berlin (1911), p. 
109; F. Konig, Cellulosechemie, 2 (1921), 109. 

3 Anon., Dinglers polytech. J., 216 (1875), 372. 

4 W. Braeutigam, Pharm. Ztg., 45 (1900), 164; Arch. Pharm., 238 (1900), 
556. 

5 M. X. Sullivan, J. Ind. Eng. Chem., 6 (1914), 919-921. 

6 O. Folin and W. Denis, J. Ind. Eng. Chem., 4 (1912), 670. 

7 E. O. von Lippmann, Ber., 37 (1904), 4521-4522. 

8 W. Ellram, Chem. Centr., II (1896), 99. 

9 R. Hegler, Flora, 73 (1890), 46. 

10 O. Richter, Z. wiss. Mikrosk., 22 (1905), 383. 

11 L. von Udranzky, Z. physiol. Chem., 12 (1888), 352-376. 



COLOR REACTIONS OF WOOD 



135 



sulphuric acid, if washed off immediately to prevent charring, 
gave a bluish green. The color reactions obtained with sul- 
phuric acid, with and without furfural, and several substances of 
possible occurrence in wood were studied. He concluded that, 
since strong acids produce furfural from wood, and since furfural 
gives a blue color with many compounds that might occur in 
wood, the reaction cannot be safely attributed to coniferin alone. 



Substance 


With furfural and 
sulphuric acid 


With sulphuric 
acid only 


Coniferin 

Vanillin 

Catechol 


Dark cherry red changing 

rapidly to violet 
Violet, a green-colored 

ring below 
Violet color changing to 

brown 
Light violet 


Same coloration 
Emerald green 
Weak green 
No color 


Gallic acid 







A step in advance was made by Nickel, 1 who attributed the 
u lignin" reactions to the aldehydic nature of wood; that these 
reactions were due to vanillin was very improbable, owing to the 
slight sensitiveness of vanillin, in comparison with wood, to phenols 
and aniline. It was found that wood behaved as an aldehyde 
towards Schiff's reagent (fuchsin + S0 2 ); wood impregnated 
with sodium bisulphite was not colored by aniline sulphate, and 
when treated with hydroxylamine did not react with phloro- 
glucinol. Similar conclusions were reached by Seliwanoff. 2 
Definite compounds are formed between lignin and phenols, and 
primary and secondary amines, but not tertiary amines; like 
an a-oxyaldehyde or a-oxy ketone, wood reacts with phenyl- 
hydrazine with the evolution of ammonia. 

In 1899, Czapek 3 isolated a substance which he named 
"hadromal" by digesting wood with strong stannous chloride 
solution and extracting the wet residue with benzene; when 
purified by conversion into the bisulphite compound and several 

1 E. Nickel, Chem. Ztg., 11 (1887), 1520; Botan. Centr., 38 (1889), 
753-756. 

2 T. Seliwanoff, Chem. Centr., I (1889), 549. 

3 F. Czapek, Z. physiol. Chem., 27 (1899), 141-166. 



136 CHEMISTRY OF CELLULOSE AND WOOD 

crystallizations from petroleum ether, it formed a light-brown 
crystalline powder melting at 75 to 80°. It was not volatile in 
steam, but was sparingly soluble in hot water, and readily so in 
organic solvents. Hadromal had phenolic and aldehydic prop- 
erties and gave the various lignin reactions to a high degree. 
The wood contained 1 to 2 per cent of the compound, which was 
apparently liberated from its combination with cellulose by the 
stannous chloride. The view of a chemical combination was 
supported by the arguments that cellulose cannot be easily 
extracted from wood by cuprammonium solution and it is not 
stained blue directly by iodine reagents. 

Hadromal, according to Grafe, 1 consists of a mixture of vanillin, 
methylfurfural, and catechol; coniferin, to which is due the blue 
coloration with phenol, is also present in wood. Hadromal was 
prepared by the gentle digestion of wood with 10 per cent hydro- 
chloric acid, and also by heating with water under pressure. 
The wood after extraction with alcohol was heated with water at 
180° for 1 hour in a sealed tube from which the air had been 
removed. The sawdust after drying was extracted with benzene. 
The green-colored solution contained "hadromal" of the compo- 
sition stated above. The resin extracted with alcohol gave the 
phloroglucinol reaction, and, judging from the depth of color for 
the extracted wood, at least half of the color-producing material 
in the wood was associated with the resin. The latter on diges- 
tion with concentrated hydrochloric acid became first yellow, 
then green like wood. The green color obtained on treating 
wood with concentrated hydrochloric acid or preferably hydro- 
bromic acid was attributed to a condensation product of 
methylfurfural with coniferin, 2 though methylfurfural has never 
been detected in wood. 

It should be mentioned that the " wound" resin (Ueberwallungs- 
harz) of Pinus laricio Poir. was found by Bamberger 3 to give a 
strong color reaction with phloroglucinol. The resin contained 
dihydrocinnamic acid, caffeic acid, ferulic acid, and a small 
amount of vanillin, the two latter compounds at least being 
colored red by phloroglucinol. Ferulic acid gives a red 

1 V. Grafe, Monatsh., 25 (1904), 987-1029. 

2 V. Grafe, I.e., p. 1029. 

3 M. Bamberger, Monatsh., 12 (1891), 445; 15 (1894), 505. 



COLOR REACTIONS OF WOOD 137 

color very similar to that obtained with lignin. 1 Besides 
vanillin, many cyclic and aliphatic aldehydes, 2 as well as ethereal 
oils or compounds containing allyl 3 groups, give a red coloration 
with phloroglucinol. 

Czapek 4 states that he has not found it possible to obtain a 
mixture of catechol, vanillin, and methylfurfural having the 
properties of hadromal; by the methods of obtaining hadromal 
employed by Grafe, oxidation may have taken place with the 
formation of vanillin and catechol. Small amounts of hadromal 
may be obtained by alcoholic extraction. On chemical and bio- 
logical grounds, hadromal is probably related to coniferyl alcohol. 

It was found by Potter 5 that boiling water or even cold water 
acting on wood for a long time extracted a substance giving the 
reactions of lignin with phloroglucinol and thalline sulphate. 
The extract obtained by heating Douglas fir and western hemlock 
sawdust with water in an autoclave gave only a faint pink with 
the phloroglucinol reagent, but the ether extract of the aqueous 
solution gave a decided red color. 6 Tests for coniferin with 
phenol and hydrochloric acid were negative. 

The violet color obtained by von Hohnel by treating cherry 
wood with hydrochloric acid, and considered by Wiesneras due 
to the presence of phloroglucinol, is without doubt due to tannins. 
Von Wagner 7 found that coniferous wood was also colored violet 
by the acid if exposed to sunlight. Definite proof of the 
occurrence of phloroglucinol in plants appears to be wanting. 8 
Votocek 9 states that the tannins in tea extract behave like phloro- 
glucinol in that they give with furfural and dilute hydrochloric 
acid similar greenish-black, water-insoluble condensation prod- 
ucts; with sufficiently concentrated acid, the extract colors wood 
a beautiful violet. Tannin extracts from willow, birch, oak, and 

1 F. Czapek, Z. physiol. Chem., 27 (1899), 151; O. Tunmann, Ber. botan. 
Ges., 30 (1912), 249. 

2 C. Hartwich and M. Winckel, Arch. Pharm., 242 (1904), 462. 

3 K. Robert, Z. anal. Chem., 46 (1907), 711. 

4 F. Czapek, "Biochemie der Pflanzen," 2nd ed., I (1913), p. 691. 

5 M. C. Potter, Ann. Botany, 18 (1904), 121. 

6 H. Schmitz, Ann. Missouri Botan. Gardens, 6 (1919), 104. 

7 R. von Wagner, Dinglers polytech. J., 228 (1878), 173. 

8 H. Moller, Ber. pharm. Ges., 7 (1897), 344-352. 

9 E. Votocek, Chem. Ztg., 37 (1913), 897. 



138 CHEMISTRY OF CELLULOSE AND WOOD 

other barks gave the same reaction. The violet color produced 
by various tannins, when a pine splinter is moistened with hydro- 
chloric acid, has been explained through the presence of vanillin in 
the wood and phloroglucinol in the tannins. : Lauffmann, 2 from a 
series of experiments, concludes that the reaction is due neither to 
the presence of phloroglucinol in tannins nor to vanillin in wood. 

Phloroglucinol is a delicate reagent for lignin ; a drop of phloro- 
glucinol solution as dilute as 1 part in 30,000 will produce a red 
spot on newsprint in 1 minute. 3 It is not infallible, however; 
Hancock and Dahl 4 found that the pith-like stem of JEschyno- 
mene aspera, though having the morphological characters of a 
true wood, gave only the faintest colorations with aniline salts 
and phloroglucinol. The wood contained 2.9 per cent of meth- 
oxyl, CH 3 0, and gave 11.6 per cent of furfural. On chlorination, 
a yellow chloride was obtained that gave the characteristic 
deeply colored solution with sodium sulphite. Von Faber 5 
found numerous instances where plant tissues were stained by 
phloroglucinol and not by the Maule reaction, and vice versa. 
The former was considered to be a specific reagent for hadromal 
and the latter for lignin. Lignified fibers, particularly jute, 
after cooking with caustic soda and bleaching, frequently fail 
to be stained by amines and phenols, though the presence of 
lignin may be shown by the Maule or the Cross and Bevan 
reaction. Renker 6 accordingly considers the latter as charac- 
teristic lignocellulose reactions, while the amines and phe- 
nols react only with traces of aldehydes in the fibers. Dark- 
brown, lignified tissues may sometimes mask the red color of 
phloroglucinol. 7 

Ground wood was found by Cross 8 and his associates to develop 
a maximum intensity of color when treated with 1 per cent of 
phloroglucinol in comparison with 10 per cent. Since the wood 

1 H. R. Procter, "Leather Industries Laboratory Book" (1908), 152, 
158-163. 

2 R. Lauffmann, Ledertech. Rundschau, 11 (1919), 61-63. 

3 C. F. Cross, E. J. Bevan, and J. F. Briggs, Ber., 40 (1907), 3121. 

4 W. C. Hancock and O. W. Dahl, Chem. News, 72 (1895), 16-18. 
s F. C. von Faber, Ber. botan. Ges., 22 (1904), 177-182. 

6 M. Renker, Papier-Fabr., (1910), Fest u. Auslandsheft, 42. 

7 C. van Wisselingh, Pharm. Weekblad, 57 (1920), 77. 

8 C. F. Cross, E. J. Bevan, and J. F. Briggs, Ber., 40 (1907), 3119-3126. 



COLOR REACTIONS OF WOOD 139 

absorbed 6.7 per cent of phloroglucinol, it would appear that 
the color development represented but a small part of the reac- 
tion, and that, in the main, there is formed a definite lignin- 
phloroglucide compound that is colorless. Zechmeister 1 came 
to the conclusion that the lignin reaction is due to a small amount 
of an aromatic aldehyde combined with a larger group (carbo- 
hydrate); the coloration with aniline salts is quinonoid. The 
aldehyde probably does not exist in wood in amounts greater 
than a few hundredths of a per cent. 2 A very similar aldehyde, 
judging from spectroscopic investigations, occurs in clove and 
sassafras oils and is probably produced from lignified tissue dur- 
ing the distillation of the oils. 

Ihl 3 thought that cinnamic aldehyde might be present in 
wood, since this aldehyde gave a yellow color with aniline sul- 
phate, a red color with phloroglucinol, and behaved like lignin 
with most reagents. The differences in color obtained with 
different woods using hydrazine sulphate, and the variations 
from the colors given by cinnamic aldehyde and vanillin, led 
Nickel 4 to point out the unsoundness of drawing conclusions 
from color reactions; piperonal, p-hydroxybenzaldehyde, and 
salicylic aldehyde gave similar colorations. According to 
Wocker, 5 vanillin or a related aromatic oxyaldehyde is present 
in lignin, since wood and vanillin give the same intensive yellow 
with benzididine chloride. The reaction takes place with lignin 
with great rapidity. Hydroxy furfural 6 or a closely related body 
has also been fixed upon as the active agent. The assumption 
of Klason 7 that coniferyl aldehyde is responsible for the color 
reactions has only a hypothetical basis. 

The intense and rapid colorations obtained with such com- 
pounds as dimethyl-p-phenylenediamine and benzidine may be 
due to the peroxidic property of wood. The action of wood on 

1 L. Zechmeister, "Zur Kenntnis der Cellulose und des Lignins," Diss. 
Zurich, (1913), p. 36. 

2 E. Crocker, I.e. 

3 A. Ihl, Chem. Ztg., 13 (1889), 560; 15 (1891), 201. 

4 E. Nickel, Chem. Ztg. 17 (1893), 1243. 

5 G. Wocker, "Die Katalyse," 2 (1915), 567. 

6 C. F. Cross, E. J. Bevan and T. Heiberg, J. Chem. Soc., 75 (1899), 
752; Ber., 40 (1907), 3120. 

7 P. Klason, Ber., 53 (1920), 711; cf. E. C. Crocker, I.e. 



140 CHEMISTRY OF CELLULOSE AND WOOD 

a photographic plate in the dark has been attributed by Russell 1 
to the liberation of hydrogen peroxide. 

In order to isolate the color-producing bodies, Wichelhaus and 
Lange 2 subjected coniferous wood to superheated steam at a 
temperature of 180°. Extraction of the distillate with ether 
yielded an oil, giving with lignin reagents, color reactions appar- 
ently identical with those obtained with wood. The distillate 
from 1200 grams of wood was separated into two fractions; 3 
the first and largest fraction gave a green precipitate with phloro- 
glucinol-hydrochloric acid, while the second gave a typical 
cherry-red precipitate. Paranitrophenylhydrazine gave a brick- 
red substance, CisHigOyNe, apparently a derivative of a keto- 
furfural formed from a hexose: 

C 6 H 12 6 + - H 2 = C 6 H 8 5 + H 2 0. 
According to a further paper, 4 two fractions were obtained 
boiling at 88 and at 95 to 105°, respectively, at 0.4-mm. pressure; 
from analyses the fractions were assigned the formulae C16H22O9 
and C16H22O10, and considered as compounds related to brasilin, 
Gi 6 Hi 4 5 , and hematoxylin, Ci 6 Hi 4 6 , which occur in dye woods. 
The color-producing constituents can be obtained by heating 
the wood with dry hydrogen chloride, removing the latter with 
an inert gas, and extracting with alcohol. 5 

Wood reacts but slowly with fuchsin-sulphurous acid solution. 
Ungar 6 obtained a yellowish-red coloration with spruce after 1 
hour and an intense reddish violet after 18 hours. 

It may be concluded that most of the so-called lignin reactions 
are due to bodies of an aldehydic nature, the structures of which 
are still problematical. The most reliable test for lignin is the 
purplish-red color of the chlorine-sodium sulphite reaction. 
The crude cellulose obtained by the cuprammonium extraction of 
wood, when treated with 72 per cent sulphuric acid to remove the 
carbohydrates, left a residue giving this characteristic color 
test for lignin. 7 

1 W. J. Russell, Proc. Roy. Soc, 61 (1897), 424; 74 (1904), 131. 

2 H. Wichelhaus and M. Lange, Ber., 49 (1916), 2001. 

3 H. Wichelhaus and M. Lange, Ber., 50 (1917), 1683-1685. 

4 H. Wichelhaus, Ber., 52 (1919), 2054-2056. 
6 H. Wichelhaus, Chem. Ztg., 47 (1923), 865. 

6 E. Ungar, Diss. Budapest (1916), p. 48. 

7 A. W. Schorger, /. Ind. Eng. Chem., 15 (1923), 812. 



CHAPTER V 
HEMICELLULOSES AND WOOD CELLULOSE 

There is no unanimity of opinion regarding definitions for 
either a cellulose or a hemicellulose. In view of the several 
criterions available, a given carbohydrate might be placed in 
either class. Since it is impossible to discuss the hemicelluloses 
without taking wood cellulose into consideration, the two will be 
treated in sequence. 

A hemicellulose may be broadly defined as a polysaccharide 
soluble in dilute alkalis and convertible into simple sugars by 
heating with dilute acids at atmospheric pressure. Until more 
data are available, the writer would add the further limitation 
that a hemicellulose, in the natural state, should be insoluble in 
boiling water. Unless this is done, the definition would include 
such totally dissimilar substances as starch that is soluble only 
in hot water; the galactan of western larch that is readily soluble 
in both cold and hot water; and the mannan of ivory nuts that is 
soluble in neither. 

Wood gum after isolation with alkalis is soluble in hot water, 
but little xylan, araban, or mannan can be isolated from wood 
with boiling water alone. On the other hand, there is reason to 
believe that all the galactan, in coniferous wood, at least, can be 
completely extracted with water. Many of the hemicelluloses, 
once peptized with alkali, remain soluble in water, but there are 
exceptions. The mannan of ivory nuts, liberated from its copper 
compound with acid, is readily soluble in hot water, but on boiling 
the solution it separates in an insoluble form. 1 Other carbo- 
hydrates become insoluble in water by being completely 
dehydrated. 

Carbohydrates such as starch, lichenin, and larch galactan 
should be placed in a distinct class, hemisaccharides. 

1 J. L. Baker and T. H. Pope, /. Chem, Soc, 77 (1900), 703. 

141 



142 CHEMISTRY OF CELLULOSE AND WOOD 

The term "hemicellulose" was introduced by Schulze. 1 He 
denned as hemicelluloses those carbohydrates occurring in the 
plant cell that are dissolved by dilute alkalis, readily hydrolyzed 
by hot dilute mineral acids, and destroyed by F. Schulze's reagent. 
(Cellulose, comparatively speaking, is resistant to these reagents.) 
If cellulose were retained as a group name, it must be qualified; 
thus a gluco-cellulose would give only glucose on hydrolysis, while 
a manno-cellulose would give mannose in addition to glucose. 

The hemicelluloses are anhydrides of hexose and pentose 
sugars. Xylan, araban, mannan, and galactan, as the names 
imply, give xylose, arabinose, mannose, and galactose, respec- 
tively, on hydrolysis. A hemicellulose giving both galactose 
and arabinose is a galacto-araban. Whether a compound of the 
latter type is a chemical individual or an intimate mixture of 
galactan and araban is not known. 

It was realized by Schulze 2 that the action of hot dilute acids 
does not furnish a sharp distinction between a cellulose and a 
hemicellulose. Pelouze, 3 in 1859, pointed out that, contrary to 
the general opinion, a boiling dilute mineral acid acting on paper, 
linen, and sawdust produces some glucose. 

The hemicelluloses were formerly known as reserve celluloses, 
the theory being that they were stored against a period of great 
metabolic activity, such as occurs in the germination of seeds 4 
and the renewal of growth in woody plants in spring. It is 
true that in seeds some of the hemicelluloses, those occurring in 
the shell 5 excepted, are consumed in growth. 6 The pentosans in 
barley do not decrease during the malting process. 7 During the 
germination of seeds there may be an increase 8 or a decrease 9 of 
pentosans. Some investigators 8 ' 9 - 10 believe that the pentosans 

IE. Schulze, Ber., 24 (1891), 2285; Z. physiol. Chem., 16 (1892), 391; 
14 (1890), 227. 

2 E. Schulze, Chem. Ztg., 19 (1894), 1465. 

3 J. Pelouze, Compt. rend., 48 (1859), 327. 

4 R. Riess, Landw. Jahrb., 18 (1889), 711. 

5 N. Castoro, Z. physiol. Chem., 49 (1906), 107. 

6 E. Schulze and N. Castoro, Z. physiol. Chem., 37 (1902), 40. 

7 B. Tollens and H. Glaubitz, J. Landw., 45 (1897), 106. 

8 A. Schone and B. Tollens, J. Landw., 48 (1900), 349; C. F. Cross, 
E. J. Bevan, and C. Beadle, Ber., 27 (1894), 1064. 

9 G. de Chalmot, Am. Chem. J., 15 (1893), 276; 16 (1894), 589. 

10 K. Goetze and T. Pfeiffer, Landw. Vers.-Sta., 47 (1896), 59. 



HEMICELLULOSES AND WOOD CELLULOSE 143 

do not serve as reserve materials, while others 1 hold that they may 
be consumed when the more readily utilizable carbohydrates are 
exhausted. 

The leaves of growing plants contain soluble pentoses. 
De Chalmot 2 found that in oak leaves the quantity of soluble 
pentoses is greater in the evening than in the morning. During 
the night they are converted into the insoluble form. Ravenna 
and Cereser 3 observed in the leaves of beans an increase in pento- 
sans during the day. When carbohydrates were administered as 
food, dextrose especially produced a marked increase in 
pentosans, from which it was concluded that the simple sugars 
are the important factors in their formation. 

During the growth of the corn plant there is a steady increase 
in pentosans. From a percentage of 7.4 in the kernel, it increases 
to 31.8 per cent in the cob at maturity. The free pentoses (as 
xylose) in the plant attained 1.66 per cent during the silk-forming 
stage. 4 

According to du Sablon, 5 hemicelluloses are formed in willow 
twigs in the fall and dissolved in spring. Storer 6 found that the 
gray birch (Betula populifolia Aiton) contained more pentosans 
in May than in October, but realized that the method of analysis 
rendered the results inconclusive. His results are too high. 
Schellenberg 7 examined twigs from various species of trees and 
could find no evidence of solution of the hemicelluloses, except 
possibly in the libriform fibers of Robinia pseudacacia. 

The pentosans in the case of some plants at least are hydro- 
philic colloids. They are mucilagenous and have the power of 
swelling and taking up an enormous amount of water, properties 
possessed by the hexosans to a far more limited extent. Succu- 

*E. Schulze, Z. physiol. Chem., 21 (1896), 392; K. Miyake, J. Coll. 
Agr. Tohoku Imp. Univ., 4 (1912), 327; C. Ravenna and O. Cereser, 
Atti R. Acad. Lincei, 18, II (1909), 177; 19, II (1910), 202; W. E. Totting- 
ham et al., J. Biol. Chem., 15 (1921), 407. 

2 G. de Chalmot, Am. Chem. J., 15 (1893), 21. 

3 C. Ravenna and O. Cereser, I.e. 

4 J. H. Ver Hulst, W. H. Peterson, and E. B. Fred, J. Agr. Research, 
23 (1923), 655-663. 

5 L. du Sablon, Rev. gen. botan., 16 (1904), 341. 

6 F. H. Storer, Bull. Bussey Inst, 2 (1897), 387. 

7 H. C. Schellenberg, Ber. botan. Ges., 23 (1905), 36- 



144 CHEMISTRY OF CELLULOSE AND WOOD 

lent plants possessing pentosans are able to take up and store 
large quantities of water that become available during dry 
periods. x 

Frank and Temme 2 thought that the wood gum in deciduous 
trees, like the resin in conifers, had a pathological significance 
and served to protect the wood beneath a wound from 
atmospheric influences. This might be the case with a true gum 
such as is secreted by Prunus, but not with "wood gum." 

If pentosans have a special function 3 in nature, no theory of 
general applicability has yet been advanced. They appear to 
be structural and not reserve materials in wood where they serve 
as a cement to hold the fibers together. 

De Chalmot 4 could find no relation between the pentosan 
content and the hardness, toughness, and other properties of the 
wood. While pentosans may play some part in the formation 
of wood, large amounts are not essential, as shown in the case 
of hemlock (Tsuga canadensis), which contains but 6 per cent. 
There was variability in the amount of pentosans between 
trees of the same species and between specimens of wood of the 
same age taken from different portions of the same tree. Age 
was of slight influence; the immature wood of Pinus tasda, 
however, contained 7.65 per cent of pentosans in comparison 
with 9.85 per cent for the mature wood. 5 

Tollens 6 thought that pentosans were formed by atmospheric 
oxidation of starch or cellulose with the aid of ferments. A 
similar suggestion was made by de Chalmot. 7 Pentoses are 
formed from hexoses by oxidation of the terminal primary alcohol 
group and the splitting out of carbon dioxide from the cor- 

1 H. A. Spoehr, "The Carbohydrate Economy of Cacti," Carnegie Inst. 
Pub., 287, 79. Washington. 

2 B. Frank and F. Temme, Biedermann's Centr., 14 (1885), 285-286. 

3 Science advances by conservative speculation, butteleological deductions, 
owing to their intriguing nature, should be accepted with caution. St. 
Pierre was so impressed by the economy of nature that in his eyes melons 
were ribbed that they might be carved and eaten more readily. At the 
opposite extreme lies Bacon's dictum, " Causarum finalium inquisito sterilis 
est, et tanquam virgo Deo consecrata, nihil parit." 

4 G. de Chalmot, Am. Chem. J., 16 (1894), 218. 

5 G. de Chalmot, Am. Chem. J., 16 (1894), 611. 

6 B. Tollens, J. Landw., 44 (1896), 187. 

7 G. de Chalmot, Am. Chem. J., 16 (1893), 610. 



HEMICELLULOSES AND WOOD CELLULOSE 145 

responding acid. The aldehyde groups are condensed so as to 
be protected from oxidation, so that pentosans result. Glucu- 
ronic acid was suggested as an intermediate product. 

Spoehr 1 has actually isolated 0.1 per cent of the lactone of 
glucuronic acid from cactus gum. He believes that under the 
influence of sunlight glucuronic acid breaks up into carbon 
dioxide and Z-xylose. It is an old observation that d-glucose and 
Z-xylose are associated in nature, but if the pentoses were formed 
by oxidation of the hexoses, d-glucose, according to Spoehr, 
would give d-arabinose. This is correct if oxidation is assumed 
to take place at the aldehyde group, but Z-xylose would be 
formed if oxidation took place at the primary alcohol group. 

The hemicelluloses, just as cotton cellulose, are soluble in 
cuprammonium solution. 2 In the natural state their solubility 
may be very slight; however, treatment with very dilute, boiling 
hydrochloric acid for 5 minutes, or cold 10 per cent acid for 24 
hours, renders them completely soluble in cuprammonium 
solution. 3 

When birch wood (Betula lutea) was ground in a ball mill with 
cuprammonium solution, the pentosans were dissolved more 
rapidly than the cellulose, though their removal was by no means 
complete. 4 

Wood gum dissolved in cuprammonium solution differs from 
cellulose in the same solvent in that it is not precipitated by 
acids until alcohol is added. 5 In case of a mixture the cellulose 
would be expected to carry down a large portion of the wood 
gum. According to Salkowski, 6 both xylan and cellulose are 
precipitated from their cuprammonium solutions by sodium 
hydroxide and cannot be separated in this way. 

The hemicelluloses are completely soluble in dilute caustic 
alkalis. 7 A cold or hot 5 per cent solution of sodium hydroxide 
is usually employed for their extraction. Isolated hemicellu- 

1 H. A. Spoehr, I.e., 42, 75. 

2 R. Riess, Landw. Jahrb., 18 (1889), 747; E. Schulze et ah, Z. physiol. 
Chem., 14 (1890), 245, 266; Ber., 22 (1889), 1194. 

3 E. Schulze, Z. physiol. Chem., 16 (1892), 410. 

4 A. W. Schorger, J. Ind. Eng. Chem., 16 (1924), 144. 

5 F. Koch, Pharm. Z. Russl, 25 (1886), 657. 

6 E. Salkowski, Z. physiol. Chem., 34 (1901), 180. 

7 E. Schulze, Z. physiol. Chem., 16 (1892), 409. 



146 



CHEMISTRY OF CELLULOSE AND WOOD 



loses are also soluble in a hot, saturated solution of sodium 
carbonate. 1 Wood gum may be isolated with lime water but 
not so conveniently as with the caustic alkalis. 2 

According to Hoffmeister, 3 pentosans are not always readily 
dissolved by caustic soda, while hexosans are frequently readily 
so. A 5 per cent solution of caustic soda extracted xylan very 
slowly from beech-wood cellulose; a 10 per cent solution removed 
it incompletely, the residual cellulose giving 1.98 per cent of 
furfural. 4 

Pine wood felled in spring was extracted with ether. Alter- 
nate extraction with water, or 0.3 per cent acetic acid, and alcohol 
gave 13.9 per cent of extract. 5 Wood felled in winter gave only 
one-seventh as much extract. The yield of furfural from the 
extract showed about 23 per cent of xylose. 

Pentosans were removed slowly and incompletely when aspen 
wood was ground in a ball mill with dilute sodium hydroxide 
for three periods of 48, 24, and 18 hours, respectively. 6 The 
amount of pentosans removed depends on the strength of the 
alkalis, though not proportionally, as shown by the following table : 





Extraction 
No. 


Wood 
dissolved, 
per cent 


Pentosan in residue 


Strength of NaOH 
solution, per cent 


Per cent 


Per cent 

of total 

pentosans 


• i 

• ( 

■ i 


1 
2 
3 

1 
2 
3 

1 
2 
3 


31.29 
3.38 
1.68 

27.82 
2.52 
1.71 

24.64 
3.57 
1.69 


8.62 
6.82 
6.71 

11.58 
9.67 
9.23 

16.01 
13.58 
12.48 


25.82 
19.42 
18.62 

36.44 
29.37 
27.34 

52.60 
42.50 
38.14 



1 M. H. O'Dwyer, Biochem. J., 17 (1893), 50. 

2 E. R. Flint and B. Tollens, Landw. Vers.-Sta., 42 (1893), 404. 

3 W. Hoffmeister, Landw. Vers.-Sta., 50 (1898), 360. 

4 M. H. O'Dwyer, Biochem. J., 17 (1923), 501-509. 

5 S. Schmidt-Nielsen, Chem. Zeit. Rep., 45 (1921), 32. 
5 A. W. Schorger, J. Ind. Eng. Chem., 16 (1924), 142. 



HEMICELLULOSES AND WOOD CELLULOSE 147 

It should be noted that sodium hydroxide as dilute as 1 per cent 
removed 61.86 per cent of the total pentosans. 

The amount of xylan remaining in straw cellulose isolated by 
chlorination and extraction with alkali at 70° depends upon the 
strength of the alkali used. 1 With a 1 per cent solution of 
sodium hydroxide, the cellulose gave 13.3 per cent furfural, and 
with a 3 per cent solution, 9.3 per cent furfural. Until the lignin 
is removed from the straw, alkali will not extract all the xylan, 
indicating a close association between pentosans and lignin. 

Schwalbe' 2 removes the pentosans from wood pulp by heating 
it with magnesium sulphite in the presence of calcium or 
magnesium oxide. 

Extraction with alkalis alters the hemicelluloses. 3 The 
xylan remaining in beech wood, after removal of the lignin, was 
not affected by chlorine dioxide; it was attacked, however, after 
isolation with 5 per cent sodium hydroxide. 4 The same was 
true of the mannan of ivory nuts. 

The xylan in straw is very slightly attacked by chlorine. 5 
Xylan, isolated with alkali, after being in contact with chlorine 
for 15 hours retained 16.8 per cent of chlorine as xylan chloride, 
while an equivalent amount of hydrogen chloride was formed. 
This indicates that chlorine attacks xylan by substitution. 
According to Briggs, 6 the presence of chlorine is due more prob- 
ably to the contamination of the xylan with lignin. In the 
case of esparto, at least, alkali simultaneously extracts a portion 
of the lignin, and this is separable from xylan only by boiling 
the extract. 

Xylan or wood gum, once isolated, is partially or completely 
soluble in boiling water. 7 Hemicelluloses may be isolated from 
wood by heating it with water 8 under a pressure corresponding 

1 E. Heuser and A. Haug, Z. angew. Chem., 31 (1918), 166. 

2 C. G. Schwalbe, G. P. 378260 (1921). 

3 J. S. Alexandrowicz, Arch. Physiol., 150 (1913), 80. 

4 E. Schmidt and E. Graumann, Ber., 54 (1921), 1867. 

5 E. Heuser and A. Haug, Z. angew. Chem., 31 (1918), 104. 

6 J. F. Briggs, Ann. Rep. Soc. Chem. Ind., 3 (1918), 128. 

7 E. Salkowski, Z. physiol. Chem., 117 (1921), 48; M. H. O'Dwyer, 
I.e., 501. 

8 K. Fromherz, Z. physiol. Chem., 50 (1906), 209; W. Koch, Diss. 
Freiburg (1909).. 



148 CHEMISTRY OF CELLULOSE AND WOOD 

to a temperature of 180°. They are also soluble in glycerine 1 
at a temperature of 210°. 

Characteristic of the pentoses and their poly anhydrides is 
the formation of furfural by dehydration when heated with 
dilute mineral acids. Stenhouse 2 appears to have been the first 
to prepare furfural by distilling with a mineral acid, in this case 

HO.HC CH.OH HC CH 

I I — > ' II II 

HO.H 2 C CH(OH).CHO HC C.CHO + 4H 2 0. 



O 

Pentose Furfural 

sulphuric acid, and to show that it may be procured from most 
vegetable materials. The highest yields of furfural were obtained 
from the hardwoods. 3 

Furfural may be produced by heating pentosans with water 
under pressure. Corncobs 4 give 20 per cent of furfural by distil- 
lation with hydrochloric acid, while with steam the maximum 
yield is about 10 per cent. The optimum conditions are a 
reaction period of 2 hours at 180°, and a ratio of cobs to water of 
not more than 1:4. 

A large number 5 of determinations of the pentosans in woods 
has been made by means of the furfural reaction. De Chalmot 6 
examined the following American species: 

Pentosans, 1 
Species Per Cent 

Tulip tree (Liriodendron tulipifera Linn.) 19. 1 

Cucumber tree (Magnolia acuminata Linn.) 17 . 7 

Wild red cherry (Prunus pennsylvanica Linn.) 19.7 

Redbud (Cercis canadensis Linn.) 21.1 

Silver maple (Acer dasycarpum Ehr.) 22 . 1 

Red gum (Liquidambar styraciflua Linn.) 21 . 1 

American holly (Ilex opaca Ait.) 24 . 6 

Dogwood (Cornus florida Linn.) 21.6 

1 Not corrected for so-called " methylpentosans." For more recent data see Table 2' 
p. 34. 

1 M. Honig, Chem. Ztg., 14 (1890), 902. 

2 J. Stenhouse, Phil. Mag., [3] 18 (1841), 122-124. 

3 J. Stenhouse, Phil. Mag., [3] 37 (1850), 227. 

4 F. B. La Forge, J. Ind. Eng. Chem., 15 (1923), 499. 

5 B. Tollens, J. Landw., 44 (1896), 171; Z. angew. Chem., 15 (1902), 508. 

6 G. de Chalmot, Am. Chem. J., 16 (1894), 218. 



HEMICELLULOSES AND WOOD CELLULOSE 149 

Pentosans, 
Species Per Cent 

Black gum (Nyssa sylvaiica Marsh.) 20. 8 

White ash (Fraxinus americana Linn.) 17.5 

Water ash (Fraxinus platycarpa Michx.) 17.4 

Butternut (Juglans cinerea Linn.) 19.2 

Shagbark hickory (Carya alba Nutt.) 21 .2 

Willow (Salix sp.) 21 .0 

Birch (Betula sp.) 23 . 4 

Beech (Fagus ferruginea Ait.) 21 .0 

Willow oak (Quercus phellos Linn.) 21.6 

White oak (Quercus alba Linn.) 20 . 4 

Red oak (Quercus rubra Linn.) 21 .7 

Water oak (Quercus nigra Linn.) 21 .3 

White elm (Ulmus americana Linn.) 17 . 4 

Sycamore (Platanus occidentalis Linn.) 21. 6 

Red juniper (Juniperus virginiana Linn.) 10.4 

White pine (Pinus stiobus Linn.) 7.5 

Shortleaf pine (Pinus miiis Michx.) 8.8 

Hemlock (Tsuga canadensis Carr.) 6.0 

The pentosans are low, 6 to 10 per cent, in the gymnosperms, 
while uniformly high, 17 to 25 per cent, in the dicotyledons. 

Furfuroids. — It is frequently difficult to establish definitely 
the presence of pentosans in plant substances, even though they 
give furfural on distillation with 12 per cent hydrochloric acid; 
furthermore, the oxycelluloses, particularly those formed by 
oxidation with chromic acid, give appreciable quantities of furfural. 
For these reasons, Cross and Bevan, 1 in 1893, advanced the 
hypothesis that oxycelluloses or "furfuroids" exist in nature. 
Esparto and the cereal straw celluloses, e.g., were classed as 
oxycelluloses. 2 They have tenaciously defended their position. 3 

Some support was given to this view by Tollens, 4 owing to the 
fact that some substances gave furfural but only a faint pentose 
reaction, 5 i.e., a pink color on warming with phloroglucinol and 

1 C. F. Cross, E. J. Bevan, and C. Beadle, Ber., 26 (1893), 2520; 27 
(1894), 1061, 1456. 

2 C. Smith, Proc. Chem. Soc, 10 (1894), 89. 

3 C. F. Cross, E. J. Bevan, and C. Beadle, J. Am. Chem. Soc, 17 (1895), 
286; C. F. Cross, Chem. News, 71 (1895), 68; C. F. Cross and E. J. Bevan 
J. Chem. Soc, 113 (1918), 182. 

4 B. Tollens, Ann., 286 (1895), 301; H. Suringar and B. Tollens, 
Z. angew. Chem., 9 (1896), 746. 

5 B. Tollens, Ber., 29 (1896), 1202. 



150 CHEMISTRY OF CELLULOSE AND WOOD 

hydrochloric acid. De Chalmot 1 and Stone, 2 on the other hand, 
considered all the furfural yielded by natural substances to be 
derived from pentosans. 

In 1896, Cross and Bevan 3 investigated the furfural-yielding 
complex in the cereal straws. This could be split off by dissolv- 
ing the cellulose in sulphuric acid of sp. gr. 1.55 to 1.65; on the 
addition of water the cellulose was precipitated, while the fur- 
furoids remained in solution. The furfuroids gave no pentose 
reaction but yielded 38 to 42 per cent of furfural. They 
assigned to furfuroids the formula of a pentose monoformal, 


C 5 H 8 03 < n > CH 2 , that theoretically would give 44 per cent of 

furfural. 

According to Heuser 4 there is no relation between the yield of 
furfural and the amount of oxycelluose present, as indicated by 
the copper number. A straw cellulose having a copper number 
of 8.60 gave 2.09 per cent furfural; the latter value was not 
changed on increasing the copper number to 15.5 by further 
bleaching. The failure of Cross and Bevan to identify pentoses 
in straw cellulose is attributed to faulty experiment. The phloro- 
glucinol reaction is significant only in the case of aqueous 
solutions of pure pentoses, and even then great care must be 
exercised; if furfural is produced, the green color masks the rose. 

Kunz 5 states that barley chaff contains two forms of substances 
yielding furfural, the one being easily hydrolyzed, the other with 
difficulty; both, however, appear to be true pentosans. There is 
no foundation, in fact, for the pentose monoformal formula. 

In 1918, Cross and Bevan 6 returned to the subject of furfuroids. 
Their experimental data point directly to the presence of a large 
proportion of pentosans in esparto cellulose, though some support 
is found for the presence of a small amount of furfuroids. Esparto 
cellulose, isolated by cooking raw esparto at 130° with 17 per 
cent of sodium hydroxide in 3 to 4 per cent solution, gave 12.5 

1 G. de Chalmot, Ber., 27 (1894), 1489; Am. Chem. J., 15 (1893), 21. 

2 W. E. Stone, Chem. News., 71 (1895), 40. 

3 C. F. Cross, E. J. Bevan, and C. Smith, J. Chem. Soc, 69 (1896), 
804; 71 (1897), 1001; Ber., 29 (1896), 1457; J. Am. Chem. Soc., 18 (1896), 8. 

4 E. Heuser and A. Haug, Z. angew. Chem., 31 (1918), 172. 

5 E. Kunz, Biochem, Z., 74 (1916), 312-339. 

6 C. F. Cross and E. J. Bevan, J. Chem. Soc., 113 (1918), 182-187. 



HEMICELLULOSES AND WOOD CELLULOSE 



151 



per cent of furfural. From the cooking liquor was isolated a 
practically pure pentosan, since it gave 50 per cent of furfural. 
Treatment of the cellulose with alkali and acid produced selec- 
tive solution of the pentosans, but there was always a loss of 
furfural. The distribution of furfural in the a-, (3-, and 7-cellu- 
loses, determined with 17.5 per cent sodium hydroxide, was: 





Yield, 
per cent 


Furfural, 
per cent 


a-Cellulose 


84.14 
15.86 


3.36 


/3-Cellulose 




7-Cellulose 


4.12 


Total 


7.48 







Since the original cellulose contained 12.5 per cent furfural, 
there was a loss of 5 per cent by the alkaline treatment. It has 
been satisfactorily shown, however, that the furfural obtainable 
from esparto cellulose comes mainly from a pentosan yielding 
xylose on hydrolysis. 1 

When sulphite wood pulp is dissolved in strong hydrochloric 
acid, 20 per cent of the apparent pentosans are destroyed. 2 
The possibility that the substances destroyed are not true pento- 
sans is indicated by the fact that the pulp, after extraction with 
17 per cent sodium hydroxide, contains 3 per cent of pentosans 
and these are not destroyed when the extracted pulp is dissolved 
in the acid. Furthermore, the pentosans in the hemicellulose 
precipitated from the alkaline extract with alcohol can be quanti- 
tatively accounted for after solution in hydrochloric acid; hence 
the apparent pentosans are destroyed by the acid and are not 
precipitable from the alkaline solution with alcohol. 

Tottingham and Gerhardt 3 hydrolyzed the spur tissue of the 
apple tree with dilute sulphuric acid and found that most of the 
constituents yielding furfural were destroyed by fermentation 
with a Carlsberg yeast; thus they were not true pentosans. 

1 J. C. Irvine and E. L. Hirst, J. Chem. Soc., 125 (1924), 15. 

2 E. Hagglund and F. W. Klingstedt, Cellulosechemie, 5 (1924), 57. 

3 W. E. Tottingham and F. Gerhardt, J. Ind. Eng. Chem., 16 (1924), 
139. 



152 CHEMISTRY OF CELLULOSE AND WOOD 

The hydrolysate with 4 per cent sulphuric acid showed 11.9 per 
cent of pentoses before fermentation and only 2.9 per cent of 
pentoses afterwards. 

Methylpentosans. — No adequate proof exists that methyl- 
pentosans occur in wood, methylfurfural or a methylpentose 
having never been satisfactorily identified. 1 The reports of 
their presence in numerous species of woods are due to failure 
to employ precise methods of examination. Most of the deter- 
minations are based on the observation of Votocek 2 that the 
phloroglucide of methylfurfural is soluble in alcohol. It is 
most probable that the soluble phloroglucide is that of co-hydroxy- 
methylfurfural. This aldehyde 3 is always formed in small 
amount during the distillation of hexose carbohydrates with 
acids. Beech wood gave 0.50 per cent of co-hydroxymethyl- 
furfural. Most of the aldehyde is formed at the end of the 
acid distillation. Methylfurfural and co-hydroxymethylfurfural, 
unlike furfural, do not color aniline acetate paper, but give a 
pink color with an alcoholic solution of aniline. 

Other substances of unknown origin increase the solubility of 
the phloroglucide precipitate. Fraps 4 found that the acid fur- 
fural distillate from plant materials lost from 7 to 23 per cent on 
redistillation. The substances destroyed were called "fura- 
loids." A portion of the loss is without doubt due to co-hydroxy- 
methylfurfural, as this aldehyde breaks down to formic and 
lsevulinic acids. 

The spectroscopic detection of methylfurfural by the method 
of Tollens, 5 who reported the presence of methylpentosan 
in pine wood, must be used with care. A hydrochloric acid 
solution of methylfurfural phloroglucide gives absorption bands 
between the F and G lines near F, and hydroxymethylfurfural, 
absorption bands between the E and F lines near F. 6 Owing 
to the similarity of the absorption bands, it is necessary, when 

1 A. W. Schorger, J. Ind. Eng. Chem., 9 (1917;, 562; 16 (1924), 143. 

2 E. Votocek, Ber., 30 (1897), 1195. 

3 M. Cunningham and C. Doree, Biochem. J., 8 (1914), 438. 

4 G. S. Fraps, Am. Chem. J., 25 (1901), 501. 

5 J. A. Widtsoe and B. Tollens, Ber., 33 (1900), 143; K. Oshima and B. 
Tollens, Ibid., 34 (1901), 1425. 

6 K. Oshima and T. Tadokoro, J. Chem. Soc. Japan, 39 (1918), 23. 



HEMICELLULOSES AND WOOD CELLULOSE 153 

both aldehydes are present, to destroy the co-hydroxymethylfur- 
fural by redistillation in the presence of hydrochloric acid. 1 

Leaves of Betula, Fraxinus, Acer, Sorbus, Tilia, and Pinus 
gave only a weak methylpentosan reaction as determined by 
the absorbtion spectra. 2 Schirmer 3 reports 12.18 per cent 
pentosan and 10.26 per cent methylpentosan in the mucilage 
obtained from the bark of the slippery elm (Ulmus fulva) . 

The apparent methylpentosan content of wood may attain 
4.5 per cent. 4 Sebelien 5 obtained 4.70 per cent for spruce. 

Preparation of Hemicelluloses. — Before attempting the isola- 
tion of hemicelluloses, it is advisable to digest the mother 
substance, particularly if it is wood, with a 2 per cent solution of 
ammonia in the cold for one or more periods of 24 hours each. 
After removal of the ammoniacal liquor, the residue is thoroughly 
washed with water. In this way most of the tannins, resins, 
and coloring matter are removed. 

Wheeler and Tollens 6 prepared wood gum from beech by 
digesting the purified sawdust with 5 per cent sodium hydroxide 
solution at room temperature for 48 hours, and adding an equal 
volume of 95 per cent alcohol to the filtrate to precipitate the 
sodium compound of the gum. The precipitate immersed in 
alcohol was decomposed with hydrochloric acid and the gum 
washed with alcohol and ether. The method does not give a 
pure product. 

The most satisfactory method from the standpoint of purity 
is based on the observation of Salkowski 7 that most of the hemi- 
celluloses are precipitated from their alkaline solution by 
Fehling's solution. According to Browne, 8 the Salkowski 
method of purification is attended with great loss of hemicellu- 
lose in the case of sugar-cane fiber. This is evidently due to the 
fact that certain forms of araban, such as those from beets and 

1 K. Oshima and K. Kondo, J. Chem. Soc» Japan, 39 (1918), 185. 

2 P. R. Solleid, Chem. Ztg., 25 (1901), 1138. 

3 W. Schirmer, Arch. Pharm., 250 (1912), 249. 

4 A. W. SchorgEr, J. Ind. Eng. Chem., 9 (1917), 556. 

5 J. Sebelien, Chem. Ztg., 30 (1906), 401. 

6 H. Wheeler and B. Tollens, Ann., 254 (1889), 304. 

7 E. Salkowski, Ber., 27 (1894), 497. 

8 C. A. Browne, J. Am. Chem. Soc., 26 (1904), 1229. 



154 CHEMISTRY OF CELLULOSE AND WOOD 

cherry gum, are not precipitated. 1 Advantage might be taken 
of this property to separate some of the hemicelluloses from 
mixtures. 

O'Dwyer 2 extracted the hemicellulose from white oak with 
cold 4 per cent sodium hydroxide and precipitated it as the 
sodium compound with alcohol. The hemicellulose, set free 
with acetic acid, was purified by the method used by Baker and 
Pope 3 for isolating mannogalactan from the clearing nut (Strych- 
nos potatorum) and mannan from the ivory nut. The ground 
nuts are extracted with a 2 per cent solution of caustic soda for 
2 hours in a water bath, filtered through glass wool, and an 
excess of Fehling's solution added to the filtrate. The blue 
precipitate settles to a pasty mass. Excess of copper salt and 
other impurities are removed by kneading and washing with 
water. The water must be slightly alkaline, as the precipitate 
becomes soluble in pure water. Cold dilute hydrochloric or 
acetic acid is added to decompose the copper compound, the 
light-green solution thus obtained being filtered and poured 
into strong alcohol. The white precipitate is washed with cold 
and then with hot alcohol, again dissolved in water, and repre- 
cipitated with alcohol. The product is collected, dried in the 
air, then at 105°, and cooled in a vacuum. 

Salkowski 4 prepared xylan from wheat straw as follows: 
boil 100 grams of comminuted straw with 2.5 liters of 6 per 
cent caustic soda for 45 minutes. After cooling, filter through 
linen, press the residue, and allow the filtrate to stand in a tall 
cjdinder for several days to settle. The supernatant liquid is 
decanted, mixed with a liter of Fehling's solution, and gently 
heated to flocculate the precipitate. The precipitate is filtered 
on linen, washed once with water, and triturated with hydro- 
chloric acid, of a concentration of about 12 per cent, until blue 
particles are no longer visible. The pulpy mass is mixed with 
2 to 3 volumes of 93 per cent alcohol, filtered, washed first with 
50 per cent alcohol, then with strong alcohol; now let stand 



1 E. Salkowski, Z. physiol. Chem., 34 (1901), 171. 

2 M. H. O'Dwyer, I.e., 502. 

3 J. L. Baker and T. H. Pope, J. Chem. Soc., 77 (1900), 699. 

4 E. Salkowski, Z. physiol. Chem., 34 (1901), 163. 



HEMICELLULOSES AND WOOD CELLULOSE 155 

several days with absolute alcohol. The xylan is finally trit- 
urated with ether and the ether removed by gentle heat. 

A xylan free from nitrogen, araban, mannan, and galactan, 
but containing 1.25 to 2.25 per cent of ash, is obtained in a yield 
of 22 to 23 per cent. For further purification, the xylan can be 
redissolved in caustic soda and the precipitation with Fehling's 
solution repeated. A cuprammonium solution may be used, but 
the precipitate is more gelatinous and difficult to handle. 

A purer product is obtained 1 by diluting the alkaline filtrate to 
14 liters and allowing to settle for several days. The solution is 
filtered, evaporated to the original volume, and again filtered 
through asbestos. The precipitate with Fehling's solution is 
dissolved by heating with ammonium hydroxide and the solution 
filtered. The filtrate is strongly acidified with hydrochloric acid 
and 1.5 volumes of 95 per cent alcohol added to complete the 
precipitation of the xylan. 

Heuser 2 starts with bleached straw pulp and obtains a xylan 
with only 0.35 per cent ash by a modification of Salkowski's 
method. The blue precipitate is kneaded twice with 80 per cent 
alcohol and filtered. The residue is then triturated with 95 per 
cent alcohol until all lumps disappear. Hydrogen chloride is 
then passed into the alcoholic suspension until the precipitate is 
pure white, a procedure requiring 3 to 4 hours. The precipitate 
is filtered off, washed repeatedly with 30 per cent alcohol, finally 
with ether, and dried. The xylan is obtained as a loose, white, 
odorless powder of a purity of 96 per cent. The yield is 19 per 
cent. 

A less pure xylan (93.8 per cent) is obtained by mixing the 
caustic soda solution with an equal volume of 96 per cent alcohol 
and passing in hydrogen chloride. 3 

Hydrolysis of Hemicelluloses. — While wood gum yields mainly 
xylose on hydrolysis, other sugars, including hexoses, are also 
obtained. O'Dwyer 4 obtained 51.5 per cent xylose, 18.5 per cent 
arabinose, and 30 per cent of a mixture of galactose and mannose 
from the hemicellulose of white oak. This partially explains the 

1 E. Salkowski, Z. physipl. Chem., 117 (1921), 48. 

2 E. Heuser and M. Braden, J. prakt. Chem., 103 (1921), 69. 

3 E. Heuser and M. Braden, J. prakt. Chem., 104 (1922), 259. 

4 M. H. O'Dwyer, I.e., 503. 



156 



CHEMISTRY OF CELLULOSE AND WOOD 



low pentosan content of wood gum. Flint and Tollens 1 obtained 
the following: 





Wood gum No. 




1 


2 


3 


4 


Pentosan, per cent 


65.36 


79.05 


81.02 


56.08 



Pentosans are hydrolyzed more rapidly by hydrochloric acid 
than sulphuric acid. 2 Councler 3 boiled 15.5 grams of beech- 
wood gum with 200 cubic centimeters of water and 10 cubic 
centimeters of hydrochloric acid of sp. gr. 1.19 for 2.5 to 3 
hours, and obtained 9.63 grams of pure xylose. Saccharifica- 
tion is accompanied by the formation of humus and dextrin-like 
bodies. 

High yields of xylose from xylan are difficult to obtain, owing 
to the partial conversion of xylose to furfural; for this reason sul- 
phuric acid is preferable to hydrochloric. 4 A 67 per cent yield 
of xylose was obtained by boiling 10 grams of xylan in 400 cubic 
centimeters of 5 per cent sulphuric acid for 70 minutes; and a 50.6 
per cent yield by hydrolysis with 5 per cent hydrochloric acid for 
30 minutes. The highest yield of xylose (84 per cent) and 
absence of furfural formation is attained by heating xylan with 
45 parts of 3 per cent nitric acid for 1 hour. 5 The xylan goes 
completely into solution after heating at 100° for 1 to 2 minutes. 
Evidently, dextrins are formed, since reduction indicates 58.72 
per cent of xylose and no sugar can be isolated. 

The decomposition of the pentose sugars during the hydrolysis 
of pentosans is much greater than when the pure sugars are 
heated with acids. 6 Xylose is less stable than arabinose. After 
heating 2 grams of sugar in 50 cubic centimeters of 10 per cent 



1 E. R. Flint and B. Tollens, Ber., 25 (1892), 2912; Landw. Vers.-Sta., 
42 (1893), 381; cf. E. Krober, J. Landw., 49 (1901), 9. 

2 R. Hatjers and B. Tollens, Ber., 36 (1903), 3306. 

3 C. Councler, Chem. Ztg., 16 (1892), 1719. 

4 E. Heuser and L. Brunner, J. prakt. Chem., 104 (1922), 264-281; E. 
Heuser and E. Kurschner, Ibid., 103 (1921), 75-102. 

5 E. Heuser and G. Jayme, J. prakt. Chem., 105 (1923), 232-241. 
*E. Schulze and B. Tollens, Landw. Vers.-Sta., 40 (1892), 379. 



HEMICELLULOSES AND WOOD CELLULOSE 



157 



sulphuric acid, on a boiling water bath for 12 hours, there 
remained : 



Residual sugar, 
per cent 



Arabinose . 
Xylose 




84.54 
73.80 



Wood Gum. Xylan. — "Wood gum " is not a true gum, since in 
the natural state it does not swell and dissolve in water. It con- 
sists mainly of xylan, usually with small amounts of mannan, 
araban, and galactan. 

Poumarede and Figuier, 1 by the extraction of wood with alkali, 
obtained a substance having the properties of pectin. It was 
looked upon as identical with arabinic acid. 2 In 1879 Thomsen 3 
investigated the extractive and called it wood gum. A large 
amount of gum was obtainable from birch wood by extracting it 
with cold caustic soda, sp. gr. 1.1 to 1.19, and precipitating with 
acid; pine gave very little gum. Sawdust from birch heartwood, 
after purification by extraction with water, alcohol, ether, alcohol, 
water, ammonia, and water, gave 19.7 per cent of wood gum. 

Koch 4 also found that wood gum occurred abundantly in the 
broad-leaved trees, but sparingly in the gymnosperms. He found 
that alcohol added to the alkaline extract precipitated a 
compound containing sodium, "5C 6 Hio05:NaOH." Hydrolysis 
gave an unknown, easily crystallizable sugar that he called "wood 
sugar." It had the following properties: m.p. 145°; specific 
rotation [a] D = +23.41°; unfermentable; and formed with phenyl- 
hydrazine long yellow needles, m.p. 160°. 

It remained for Tollens 5 and his associates to clarify 
the composition of wood gum. By one extraction with sodium 
hydroxide, 1300 grams of birch sawdust gave from 63 to 70 grams 
of nearly white wood gum. A second extraction gave 14 grams 
of a dark product. The purified gum dissolved completely in 

1 J. A. Poumarede and L. Figuier, Compt. rend., 23 (1846), 918. 

2 E. Reichardt, Arch. Pharm., 7 (1877), 116. 

3 T. Thomsen, J. prakt. Chem., 127 (1879), 146-168. 

4 F. Koch, Pharm. Z. Russl, 25 (1886), 619-770. 

5 H. Wheeler and B. Tollens, Ann., 254 (1889), 304-320; Ber., 22 
(1889), 1046. 



158 



CHEMISTRY OF CELLULOSE AND WOOD 



sodium hydroxide of a concentration slightly over 2 per cent. 1 
Ultimate analysis agreed with the formula C 6 Hio0 5 ; however, 
hydrolysis yielded the second known pentose sugar, xylose, 
having [a] D = +18 to 19°. From 60 grams of gum, 16 grams of 
crystalline sugar were obtained. 

Fir wood yielded 4.0 per cent of wood gum. 2 Hydrolysis gave 
xylose having [a] D = +18.69°, and yielding an osazone of m.p. 
157.5 to 158.5°. 

Okumura 3 determined the wood gum in various species of 
Japanese woods by extraction with 5 per cent sodium hydroxide. 
The yields varied from 0.961 (Abies firma) to 19.72 per cent 
(Fagus sieboldi). 

It is of interest that, while the true gum 4 secreted from the bark 
of the cherry yields arabinose with a little xylose, the "gum" of 
the wood gives mainly xylose. 5 

Wood gum dissolved in dilute sodium hydroxide is strongly 
Z-rotatory. 



Source 


M» 


Author 


Birch 

Beech 


-92. 73 to -96.55° 
-69.62 

-81.0to -83.6 
-84.0 

-75 


Koch, I.e. 

Wheeler and Tollens, I.e. 


White oak 


Thomsen, Ber., 14 (1881), 136 
Thomsen, Ber., 13 (1880), 2168 
O'Dwyer, I.e. 







Wood gum is highly hygroscopic and is usually high in ash. 
Two samples examined by Councler 6 had the following 
comnosition : 



composition 



Source 



Ash, 
per 
cent 



Water, 
per 
cent 



Organic 
matter, 
per cent 



Beech wood. 
Straw paper . 



1.79 
1.23 



17.44 
13.36 



80.77 
85.41 



1 H. Wheeler and B. Tollens, Ann., 254 (1889), 320. 

2 H. Wheeler and B. Tollens, Ann., 254 (1889), 323. 

3 J. Okumura, Bull. Coll. Agr. Japan, 2 (1894), 76. 

4 C. A. Browne and B. Tollens, Ber., 35 (1902), 1466. 

5 E. W. Allen and B. Tollens, Ann., 260 (1890), 296. 

6 C. Councler, Chem. Ztg., 16 (1892), 1719. 



HEMICELLULOSES AND WOOD CELLULOSE 159 

Xylan from corncobs contained 13.5 per cent hygroscopic 
moisture and that from white birch (Betula alba) 9.42 to 10.26 
per cent. 1 Koch 2 heated wood gum at 110° and obtained the 
following values: 

Hygkoscopic Water, 
Source Per Cent 

Quebracho wood 15 . 34 

Oak wood 16 . 95 

Walnut shells 19.93 

Wood gum is colored by the phloroglucinol reagent only on 
warming, while lignin is colored in the cold. 3 

Pure beech-wood gum has been stated 4 to reduce a mercuric 
chloride solution on standing 4 weeks. Five grams of gum 
reduced 2.8 grams of mercuric chloride in a solution of 10 grams 
of the salt in 285 cubic centimeters of water. The purity of the 
gum is open to question. 

Pure xylan should have the formula C5H 8 4 . Judging from 
ultimate analyses, 5 the pure compound has seldom been obtained. 
Aside from non-carbohydrates, the pentosans are usually con- 
taminated with hexosans. A wood gum prepared from beech 
by the method of Wheeler and Tollens contained 2.8 per cent 
methoxyl, the original wood 5.4 per cent. 6 Pure xylan, free 
from lignin, does not give the methoxyl reaction. 7 

Konig 8 obtained succinic acid, m.p. 185°, by fusing beech-wood 
gum with caustic potash. Catechol was not formed. Straw 
xylan 9 fused with 10 parts of caustic potash at 230 to 280° gave 
about 50 per cent of oxalic acid, 13 to 15 per cent of acetic acid, 
and 8 to 12 per cent of formic acid. A slight amount of succinic 
acid could be detected. The catechol and protocatechuic acid, 

1 S. W. Johnson, J. Am. Chem. Soc., 18 (1896), 214. 

2 F. Koch, Pharm. Z. Russl, 25 (1886), 655. 

3 H. Wheeler and B. Tollens, Ann., 254 (1889), 332. 

4 A. Link and A. Voswinkel, Pharm. Zentralhalle, 31 (1890), 679. 

5 S. W. Johnson, I.e.; F. Koch, I.e., 670; H. Wheeler and B. Tollens, 
I.e., 321; M. H. O'Dwyer, I.e., 503. 

6 R. Benedict and M. Bamberger, Monatsh., 11 (1890), 262. 

7 E. Heuser and M. Braden, J. prakt. Chem., 104 (1922), 259. 

8 F. Konig. Cellulosechemie, 2 (1921), 118. 

9 E. Heuser and J. Roth, J. prakt. Chem., 107 (1924), 1-6. 



160 



CHEMISTRY OF CELLULOSE AND WOOD 



qualitatively detected, were probably formed from ligneous 
impurities. 

When xylan is dissolved in 2.5 times its weight of nitric acid 
(sp. gr. 1.2) at 70°, and the solution held at 45°, 21.7 per cent of 
the theoretical yield of trihydroxyglutaric acid is formed. 1 With 
twice the weight of nitric acid the chief product is oxalic acid. 

Birch-wood gum heated with water at 84 pounds steam 
pressure for 2 hours gave 0.13 to 0.16 per cent of acetic acid and 
0.11 to 0.16 per cent formic acid. 2 

Birch- wood gum on destructive distillation 2 gave the following: 

Per Cent 

Charcoal 37 . 2 

Tar and oil 33 . 6 

Pyroligneous acid 33 . 6 

Gas 18.1 

The pyroligneous acid contained 9.0 per cent of furfural. The 
yields of acetic and formic acids were 0.29 and 0.17 per cent, 
respectively, so that wood gum is of no importance in the forma- 
tion of acetic acid in the destructive distillation of wood. The 
wood gum from fir gave similar results. 

Heuser and Scherer 3 distilled pure xylan from straw at 
atmospheric pressure (I) and in vacuo (II). 





I 


II 


Charcoal 


31.6 

6.7 

43.8 

18.0 


31.5 


Tar 


21.4 


Pyroligneous acid 


20.1 


Gas 


27.0 







The distillates contained 7.5 and 6.6 per cent of acid calculated 
as acetic acid, though neither formic nor acetic acid could be 
detected, and 6.0 and 2.18 per cent of furfural, respectively. The 
syrupy distillate from 200 grams of xylan distilled in vacuo gave 1 



1 E. Heuser and G. Jayme, J. prakt. Chem., 105 (1923), 232-242. 

2 H. Bergstrom, Papier-Fabr., 11 (1913), 759. 

3 E. Heuser and A. Scherer, Brennstoff-Chem., 4 (1923), 97-101. 



HEMICELLULOSES AND WOOD CELLULOSE 161 

gram of a crystalline compound, C 5 H 6 03, m.p. 206.5°, that prob- 
ably represents a dehydration product of xylan. 

Esters and Ethers. — Xylan from beech and acacia woods, 
subjected to acetolysis with acetic anhydride and sulphuric 
acid, gave only amorphous products. 1 Bader 2 obtained mono- 
and diacetyl derivatives of wood gum, but more than two 
acetyl groups could not be introduced. The Schotten-Baumann 
method gave a monobenzoyl derivative as a crumbly, amorphous 
mass. A higher benzoyl derivative could not be obtained. 

Komatsu and Kashima 3 found that the xylan from wheat 
straw acetylated readily, giving mono- and diacetyl derivatives. 
In marked contrast, the xylan from corncobs could scarcely be 
acetylated by the action of acetic anhydride and acetyl chloride, 
with and without catalysts. Straw xylan had the rotation 
[a] D = —78° in a 2.5 per cent solution of caustic soda. It 
gave only a trace of xylose on heating with water at 145° for 3 
hours. The xylan from corncobs had the rotation [a] D = —80°. 

Beech-wood gum, nitrated 4 at 0° with 10 parts of nitric acid 
of sp. gr. 1.52, swelled but did not dissolve during 24 hours. It 
contained 12.31 per cent of nitrogen, thus agreeing with a dini- 
trate. The ester was insoluble in the usual solvents, such as 
acetone and ethyl acetate. 

The nitration of pure xylan from straw cellulose 5 is slight 
with an acid weaker than 

HN0 3 :H 2 S0 4 :H 2 = 25.10:66.85:8.05. 

There was obtained a yield of 115.7 per cent of xylan nitrate 
containing 8.76 per cent of nitrogen. Of this ester, 61 per cent 
was soluble in ether-alcohol mixture and contained 7.13 per cent 
of nitrogen. The remainder was soluble in ethyl and amyl 
acetates and contained 9.22 per cent of nitrogen. 



1 G. Zemplen, Z. physiol Chem., 85 (1913), 182. 

2 R. Bader, Chem. Ztg., 19 (1895), 55. 

3 S. Komatsu and K. Kashima, Mem. Coll. Sci. Kyoto Imp. Univ., 5 
(1922), 307-314. 

4 W. Will and F. Lenze, Ber., 31 (1898), 89; cf. R. Bader, I.e. 

5 B. Rassow and E. Dorr, J. prakt. Chem., 108 (1924), 144. 



162 CHEMISTRY OF CELLULOSE AND WOOD 

Xylan dissolved in sodium hydroxide will accept only 1 to 
1.5 methyl groups when treated with dimethyl sulphate. 1 The 
product partially methylated in this way is completely converted 
into the dimethyl ether, softening at 65 to 70°, by heating with 
methyl iodide and silver oxide at 100°. 

Araban. — Araban, or a carbohydrate yielding arabinose, is 
present in wood. Hauers and Tollens 2 obtained a small amount 
of arabinose by the hydrolysis of pine wood with a solution of 
calcium bisulphite at a temperature between 115 and 135°. 
Arabinose has been identified in the products of hydrolysis of 
white spruce 3 (Picea canadensis), and in the hemicellulose from 
beech 4 and American white oak. 5 According to Browne, 6 
deciduous trees contain 15 to 25 per cent of pentosans consisting 
of xylan and a small amount of araban. 

The lignin isolated from pine wood by means of strong hydro- 
chloric acid contains 4 to 6 per cent of pentosan that is considered 
as chemically combined with the lignin. According to Hagglund, 7 
the pentosan is chiefly araban, and according to Kiirschner, 8 
xylan. Heuser 9 looks upon the pentosan in lignin as accidental 
and as due to incomplete hydrolysis, not chemical combination. 

Laevulan. — The presence of fructose in the products of hydrol- 
ysis of wood has been reported on various occasions, identifica- 
tion being largely based on color reactions. The most reliable 
evidence is that furnished by Hagglund, 10 who obtained 
a-methylphenylfructosazone, m.p. 147 to 148°, from sulphite 
pulp. The pulp contained 2.5 per cent laevulan. Arabinose, 
fucose, rhamnose, and galactose were absent. Sherrard 11 did 

1 E. Heuser and W. Ruppel, Ber., 55 (1922), 2084. 

2 R. Hauers and B. Tollens, Ber., 36 (1903), 3319. 

3 E. C. Sherrard and G. W. Blanco, J. Ind. Eng. Chem., 15 (1923), 611. 

4 C. A. Browne and B. Tollens, Ber., 35 (1902), 1466. 

5 M. H. O'Dwyer, Biochem. J., 17 (1923), 501. 

6 C. A. Browne, "Handbook of Sugar Analysis," N. Y. (1912), p. 553. 
7 E. Hagglund, Ber., 56 (1923), 1866; E. Hagglund and C. J. Malm, 

Cellulosechemie, 4 (1923), 73. 

8 K. Kurschner, Naturprodukte, 24 (1923), 36. 

9 E. Heuser, Cellulosechemie, 4 (1923), 77. 

10 E. Hagglund and F. W. Klingstedt, Cellulosechemie, 5 (1924), 57. 

11 E. C. Sherrard and G, W. Blanco, J. Ind. Eng, Chem., 15 (1923), 612. 



HEMICELLULOSES AND WOOD CELLULOSE 163 

not obtain fructose on hydrolyzing white spruce (Picea 
canadensis). 

There is no satisfactory datum that wood contains a hemi- 
cellulose giving glucose on hydrolysis. 

Mannan. — In 1889, Tollens 1 obtained from wood (species not 
stated) what appeared to be mannose hydrazone. Subsequent 
investigations have shown that mannan is widely distributed 
in woods and is especially characteristic of the gymno- 
sperms. Bertrand 2 states that in the gymnosperms mannan 
occupies the place of xylan in the hardwoods. Mannose was 
obtained from the Cycadece and Coniferce, but it was absent or 
nearly so in the Gnetacew, which are not true gymnosperms. 

Tollens 3 and his associates found mannose in sulphite liquor, 
the raw material generally employed being spruce (Picea excelsa 
Lk.). Mannan occurs not only in the wood 4 of this species, but 
in the needles 5 and seeds as well. 

Mannan was found by Storer 6 in the wood of white pine 
(Pinus strobus); pitch pine (Pinus rigida); Norway spruce 
(Picea excelsa); Japanese larch (Larix leptolepis); hemlock 
(Tsuga canadensis); red cedar (Juniperus virginiana); and white 
cedar (Chamcecyparis thyoides). An extensive investigation of 
American species renders it highly probable that mannan occurs 
in all the gymnosperms. 

No regularity appears to exist in the amount of mannan 
occurring in wood when age, season, and the same and different 
species are considered. Storer 7 obtained 3.61 per cent of man- 
nose from a white pine felled in August and 1.86 per cent from 
one felled in December. It is improbable that there is a seasonal 
variation in the mature wood, though he states that pine trees 
contain much more mannan in late summer than in early winter. 

1 H. Wheeler and B. Tollens, Ber., 22 (1889), 1046. 

2 G. Bertrand, Compt. rend., 129 (1899), 1025. 

3 J. B. Lindsey and B. Tollens, Ann., 267 (1892), 349; Ber., 23 (1890), 
2990; Z. angew. Chem., 5 (1892), 155. 

4 P. Klason, "Zusammensetzung des Fichtenholzes," Berlin (1911), p. 33. 

5 J. B. Lindsey and B. Tollens, Z. angew. Chem., 5 (1892), 154. 
6 F. H. Storer, Bull. Bussey Inst., 3 (1902), 13-45, 47-69. 

7 F. H. Storer, I.e., 37, 51. 



164 CHEMISTRY OF CELLULOSE AND WOOD 

Mannan in Woods 

Mannan, 

Species Pek Cent 

Douglas fir (Pseudotsuga taxifolia Briton) 6 . 65 

Corkbark fir (Abies arizonica Merr.) 6 . 57 

Western larch (Larix occidentalis Nutt.) 5 . 13 

Arborvitse (Thuja occidentalis Linn.) 1 . 44 

White spruce (Picea canadensis Mill.) 7 . 12 

Longleaf pine (Pinus palustris Mill.) 4 . 75 

Loblolly pine (Pinus tceda Linn.) 5 . 10 

Sugar pine (Pinus lambertiana Dougl.) 4 . 67 

Coulter pine (Pinus coulteri Lamb.) 5 . 22 

Monterey pine (Pinus radiata Don.) 7 . 68 

Pinon pine (Pinus edulis Engelm.) 6 . 00 

California swamp pine (Pinus muricata Don.) 3 . 07 

Chihuahua pine (Pinus chihuahuana Engelm.) 5 . 00 

Western white pine (Pinus moniicola Dougl.) 6 . 93 

Whitebark pine (Pinus albicaulis Engl.) 4 . 22 

Western yellow pine (Pinus yonder osa scopulorum Englm.) 4.64 

Jeffrey pine (Pinus jeffreyi) 5 . 40 

Digger pine (Pinus sabiniana Dougl.) 7 . 17 

Limber pine (Pinus flexilis James) 5 . 94 

Bristlecone pine (Pinus aristata Engelm.) 5.41 

Knobcone pine (Pinus attenuata Lemm.) 3 . 57 

Cryptomeria japonica Don 6 . 35 

Redwood (Sequoia sempervirens Endl.) 3.21 

Western yellow pine (Pinus ponder osa Laws.) 6 . 37 

Sugar pine (Pinus lambertiana Dougl.) 6 . 63 

Taxus baccata Linn 9 . 10 * 

Cupressus torulosa Don 3.4 

Abies pectinata D.C 9.6 

Araucaria brasiliana A. Rich 9.5 

Pinus laricio Poir., cone 8.4 

* As mannose. 

i A. W. Schorger, J. Ind. Eng. Chem., 9 (1917), 748. 

2 C. Kimoto, Bull. Coll. Agr. Japan, 5 (1902), 254. 

3 W. H. Dore, J. Ind. Eng. Chem., 12 (1920), 476. 
« G. Bertrand, Compt. rend., 129 (1899), 1025. 

Examination 1 of a radial section of a sugar pine 4 feet in 

diameter showed that the sapwood contained 4.25 per cent 
mannan and the heart wood from 4.58 to 4.80 per cent. The 
mannan content of the sapwood in some species is higher than 
that of the heartwood: 

i A. W. Schorger, J. Ind. Eng. Chem., 9 (1917), 748. 



HE MI CELLULOSES AND WOOD CELLULOSE 



165 



Mannan content, per cent 




Nor does it follow that the youngest wood is the richer; a 
longleaf pine sapling 6 years old contained 1.64 per cent mannan 
in comparison with 4.75 per cent for a mature tree. 

No mannose was found in a sound pine log that had been buried 
in a swamp for 200 years, or in decayed and weathered pine wood 
taken from the surface. 1 

Hagglund 2 hydrolyzed spruce wood in two ways: 

I. Heated with 0.5 per cent sulphuric acid at 155 to 165° for 30 
minutes; yield of sugar as glucose 9 per cent. 

II. Heated with 0.5 per cent sulphuric acid at 170° for 45 
minutes; yield of sugar as glucose 14.5 per cent. 

Mannose was the principal sugar formed, as shown by the 
following table: 



• 


Sugar 


I, 

per cent 


II, 

per cent 


Xylose 

Mannose 

Fructose 

Galactose 

Glucose 

Undetermined 


32.7 

35.6 

4.3 

3.3 

0.5 

23.6 


30.1 

43.7 

4.7 

3.3 

14.0 

4.2 



He believed that only the hemicelluloses were hydrolyzed in 
experiment I, the cellulose being attacked with increasing time 
and temperature. 

Mannan appears to occur sparingly in some hardwoods and to 
be absent in others. The evidence shows irregularity. Mannose 
has been obtained from the hemicellulose of white oak. 3 Early 

1 F. H. Storer, Bull. Bussey Inst, 3 (1903), 69. 

2 E. Hagglund, Biochem. Z., 70 (1915), 425. 

3 M. H. O'Dwyer, I.e., 501. 



166 CHEMISTRY OF CELLULOSE AND WOOD 

papers 1 record the presence of mannose in the hydrolyzate 
from aspen (Populus tremula), but Heuser and Brotz 2 have 
recently reported its absence. It was not found by Schorger 3 
or Storer 4 by the direct hydrolysis of a number of deciduous 
woods, including Populus tremuloides; nor by Dore 6 in California 
live oak (Quercus agrifolia Nee.). 

Mannose was obtained from the inner bark of several conifers 
but not from that of Betula populifolia and Populus tremuloides. 6 

Galactan. — Koch, 7 in 1886, obtained a small amount of mucic 
acid by oxidizing wood gum with nitric acid, thus showing the 
presence of a galactose compound. Galactose was subsequently 
found in sulphite liquor, 8 the woods usually cooked being conif- 
erous. In the sulphite liquor from Pinus sylvestris, Hagglund 9 
found 0.27 per cent of galactose, based on the wood. 

Storer 10 states that he was unable to detect much, if any, 
galactan in the wood of trees. 

In 1916, it was shown by Schorger and Smith 11 that galactans 
are characteristic of the conifers and unique among wood carbo- 
hydrates in being very easily soluble in water. It remains to 
be seen if the various species of woods contain the same galactan. 

Only the e-galactan of the western larch (Larix occidentalis 
Nutt.) has been extensively studied. A gum-like excrescence 
of the western larch had been examined, but not identified, by 
Trimble. 12 The galactan content of the wood varies from 8 to 
17 per cent. To isolate it, the aqueous extract is first purified 

1 K. Fromherz, Z. physiol. Chem., 50 (1906), 237; W. Koch, Diss. Frei- 
burg (1909), 7. 

2 E. Heuser and A. Brotz, Papier-Fabr. Fest u. Auslandsheft, 23 (1925), 
69. 

3 A. W. Schorger, I.e. 

4 F. H. Storer, Bull. Bussey Inst., 3 (1902), 32, 53, 108. 
6 W. H. Dore, J. Ind. Eng. Chem., 12 (1920), 984. 

6 F. T. Dillingham, Bull. Bussey Inst., 3 (1906), 127, 128. 

7 F. Koch, Pharm. Z. Russl, 25 (1886), 668. 

8 T. Seliwanoff, Chem. Zentr., 60, I (1889), 549; J. B. Lindsey and B. 
Tollens, Z. angew. Chem., 5 (1892), 154; H. Krattse, Chem. Ind., 29 (1906), 
217. 

9 E. Hagglund, Biochem. Z., 70 (1915), 416. 

10 F. H. Storer, Bull. Bussey Inst., 3 (1902), 39. 

11 A. W. Schorger and D. F. Smith, J. Ind. Eng. Chem., 8 (1916), 494. 

12 H. Trimble, Am. J. Pharm., 70 (1898), 152. 



HEMICELLULOSES AND WOOD CELLULOSE 167 

by adding tannic acid, then an excess of lead acetate, and filtering. 
The excess lead is removed with hydrogen sulphide. On adding 
an excess of alcohol to the clear, concentrated filtrate, the galac- 
tan is precipitated as a gummy mass that soon becomes granular. 

Pure e-galactan forms a white powder that dissolves easily in 
cold water to form a mobile liquid. The rotation is [a] 2 ^° = 
+ 12.11°. The galactan gave 6.18 per cent of furfural, but no 
sugar other than galactose could be identified. Ultimate analysis 
and a molecular-weight determination by the freezing-point 
method gave the formula (C 6 Hi O 5 )20. The acetyl derivatives 
were amorphous. 

Galactan has been identified in the following hardwoods: 
aspen (Populus tremula); 1 white oak; 2 apple wood; 3 California 
live oak (Quercus agrifolia). 4 

According to Salkowski, 5 galactan is not precipitated by 
Fehling's solution and appears to be destroyed by heating with 
alkalis, even ammonia. 

Wood Cellulose. — The cotton hair is unique in consisting 
almost entirely of orthoglucosan, i.e., glucose anhydride, 6 and is 
universally recognized as the source for standard cellulose. It 
is entirely possible that celluloses exist in which mannan or 
xylan is chemically combined with the orthoglucosan. One 
school holds to the opinion that all celluloses when properly 
purified are identical with cotton cellulose ; another, that there are 
celluloses containing the anhydrides of several different sugars. 
The mode of linkage, if any, between orthoglucosan and xylan 
or mannan is problematical, though, reasoning from analogy, 
it would probably be glucosidal. 

Cellulose has been defined in various ways. In order to have 
a concrete working basis, Schorger 7 defined wood cellulose as the 
residue remaining after alternate treatment of wood with chlorine 
gas and sodium suphite to the point where the chlorine-sodium 
sulphite, or the Maule reaction for lignin, ceases. 

1 K. Fromherz, Z. physiol. Chem., 50 (1906), 238. 

2 M. H. O'Dwyer, I.e. 

3 W. E. Tottingham et 0,1., J. Biol. Chem., 21 (1920), 407. 

4 W. H. Dore, J. Ind. Eng. Chem., 12 (1920), 984. 

5 E. Salkowski, Z. physiol. Chem., 34 (1901), 175, 177. 

6 A. W. Schorger, J. Ind. Eng. Chem., 16 (1924), 141. 

7 A. W. Schorger, J. Ind. Eng. Chem., 9 (1917), 561. 



168 CHEMISTRY OF CELLULOSE AND WOOD 

Klason 1 has recently defined cellulose as the carbohydrate 
residue that does not dissolve at 98° in a solution containing 13 
grams of free and 37 grams of half -bound sulphur dioxide per 
liter. He finds that when wood is digested with sulphite liquor 
under the above conditions, the yield of cellulose remains con- 
stant, even though the digestion be prolonged with fresh liquor. 
Wood cellulose will also withstand a liquor containing 32 grams 
of sulphur dioxide, its fibrous structure being retained. Cotton 
behaves quite differently with the stronger acid. Only 90.5 
per cent of the cellulose is recovered, and this is in a structureless 
form similar to hydrocellulose. 

The term " cellulose" has been used by Cross 2 "as the com- 
prehensive expression for the non-nitrogenous basis of the vege- 
table world." This definition embraces the simple and compound 
celluloses and is perhaps satisfactory from a botanical viewpoint. 

There are many adherents to the theory that the normal 
cellulose of wood contains adsorbed hexosans and pentosans. 
The adsorption theory has been developed at length by Wisli- 
cenus 3 to account for plant growth. Just how colloids can be 
synthesized by the cell protoplasm and transferred to the cell 
wall for adsorption is not at all clear. 

According to Wise, 4 cellulose consists of small units arranged, 
for the purpose of illustration, like bricks in a wall. The units 
are held together by primary and secondary valencies. Wood 
cellulose is to be distinguished from cellulose only by the presence 
of variable amounts of adsorbed carbohydrates. 

Herzog and Jancke 5 obtained identical Rontgen-ray spectro- 
grams from cotton, ramie, linden wood, and wood pulp, indicat- 
ing that the same cellulose occurs in each. 

The opinion is held by Heuser 6 that all plant celluloses, regard- 
less of their source, are identical. Bleached coniferous cellulose 
was purified by boiling three times with 6 per cent sodium 

1 P. Klason, Papier-Fabr., 22 (1924), 373. 

2 C. F. Cross, in: " Chemistry in the Twentieth Century," N. Y. (1924) 
p. 159. 

3 H. Wislicenus and M. Kleinstuck, Kolloid Z., 6 (1910;, 17, 87. 
4 L. E. Wise, J. Ind. Eng. Chem., 15 (1923), 711. 

5 R. O. Herzog and W. Jancke, Ber., 53 (1920), 2162. 

6 E. Heuser and E. Boedeker, Z. angew. Chem., 34 (1921), 461; E. 
Heuser and A. Haug, Ibid., 31 (1918), 173. 



HEMICELLULOSES AND WOOD CELLULOSE 169 

hydroxide or, as an alternative, extracted with cold 17 per cent 
sodium hydroxide. In this way the pentosan content was 
reduced from 4.06 to 1.92 per cent. The purified cellulose and 
cotton were hydrolyzed with strong hydrochloric acid for com- 
parison; in both cases the maximum dextrose reading was obtained 
after 16.5 hours. From the wood cellulose, 94.2 per cent of 
dextrose was isolated in the pure condition. 

The identity of wood cellulose with cotton cellulose has been 
claimed 1 on the ground of having obtained comparable yields of 
cellobiose octacetate after the wood cellulose had been freed from 
pentosans with 17.5 per cent sodium hydroxide. 

The action of boiling 6 per cent, or cold 17.5 per cent sodium 
hydroxide, is too severe to permit conclusions to be drawn con- 
cerning adsorption or chemical combination. The same may be 
said of acetolysis when applied to the crude or purified wood 
cellulose. The splitting off of pentosans could readily occur with 
any of these reagents. 

It is a question whether the formation of cellobiose octacetate 
is a criterion for the identity of celluloses. The lichenin from 
Iceland moss (Cetraria islaridica) gives cellobiose octacetate, 
acetobromocellobiose, and a triacetate having a rotation ([a]i) 9 ° 
= —23.8°) very similar to cellulose triacetate; 2 and when dis- 
tilled rapidly at 12-millimeters pressure, lsevoglucosan 3 is obtained. 
Lichenin, nevertheless, is soluble in dilute alkali and gives a 
clear colloidal solution with boiling water. Furthermore, it 
hydrolyzes readily with dilute acids at atmospheric pressure, 
giving 64.86 per cent of dextrose. 4 

Lichenin from its ease of hydrolysis would, accordingly, 
correspond to Schulze's definition of a hemicellulose. Salkowski 4 
believes that the degree of hydrolysis depends on the physical 
condition of the carbohydrate; the more finely it is divided the 
more drastic is the action. This case illustrates the difficulty in 
drawing a line between a cellulose and a hemicellulose. Karrer 5 

1 L. E. Wise and W. C. Russell, J. Ind. Eng. Chem., 14 (1922), 285; 
15 (1923), 815; cf. S. V. Hintikka, Cellulosechemie, 4 (1923), 62. 

2 P. Karrer and B. Joos, Biochem. Z., 136 (1923), 537; Helvetica Chim. 
Acta, 6 (1923), 800; P. Karrer and K. Nishida, Ibid., 7 (1924), 363. 

3 P. Karrer and M. Staub, Helvetica Chim. Acta, 7 (1924), 928. 

4 E. Salkowski, Z. physiol. Chem., 114 (1921), 31. 

5 P. Karrer et al., Helvetica. Chim. Acta, 6 (1923), 800; 7 (1924), 144, 518, 



170 CHEMISTRY OF CELLULOSE AND WOOD 

considers lichenin to be a "reserve" cellulose (hemi cellulose), as 
it is quantitatively converted to dextrose by the enzyme lichenase 
found in the snail, Helix pomalia. Cotton is scarcely attacked. 
The identity of "normal" celluloses may also be questioned. 
The cellulose isolated from marine fiber (Posidonia australis) is 
regarded as consisting, like cotton, of glucose anhydride residues. 
The optical rotations of their triacetates, however, differ greatly, 
as shown by the following: 

[ah 

Cotton triacetate -22.3° 

Posidonia triacetate —39 . 8° 

This variation indicates a difference in chemical structure. 1 

Esparto cellulose, according to Irvine and Hirst, 2 behaves as 
a mixture or solid solution containing 81.5 per cent normal 
cellulose and 18.5 per cent xylan. On being freed from xylan, 
esparto cellulose acts like cotton cellulose, giving cellobiose 
octacetate and 2, 3, 6-trimethylglucose. It is noteworthy that 
in no case was cellobiose octacetate obtained from the cellulose 
that had not been freed from xylan by alkaline extraction. 

The approximately complete elimination of pentosans from 
wood or straw cellulose is accompanied by a considerable loss of 
cellulose due to the persistence with which the pentosans are 
retained. Straw cellulose 3 isolated by the chlorine method gave 
10.30 per cent furfural (22.34 per cent xylan). After six extrac- 
tions with hot 6 per cent sodium hydroxide, the residue gave 1.97 
per cent furfural. Heuser found it difficult, if not impossible, to 
free the cellulose completely from xylan. Nor could the xylan 
be eliminated by treating the straw directly with sodium 
hydroxide; after five treatments the residue (40 per cent) con- 
tained 9.69 per cent xylan. Treatment of the residue with 
chlorine did not reduce the amount of xylan. 

Heuser 4 remarks that nearly complete elimination of pentosans 
would be expected through digesting straw with caustic soda at a 

i J. C. Earl, J. Chem. Soc., 125 (1924), 1322. 

2 J. C. Irvine and E. L. Hirst, J. Chem. Soc, 125 (1924), 15-25; cf. 
E. L. Hirst, J. Soc. Chem. Ind., 41 (1922), 392R. 

3 E. Heuser and A. Haug, Z. angew. Chem., 31 (1918), 166-168. 

4 E. Heuser, I.e., 168. 



HEMICELLULOSES AND WOOD CELLULOSE 171 

pressure corresponding to 140 to 150°. It is a surprising fact, 
however, that the straw pulp retains 29 per cent of xylan. 

In the case of sulphite wood pulp, after two extractions with 17 
per cent sodium hydroxide, the pentosans were distributed as 
follows : 

Pentosans, 
Per Cent 

a-Cellulose 0.8 

/3-Cellulose 1.26 

7-Cellulose 2.16 

There is, accordingly, no way of telling if the hemicelluloses 
and cellulose are combined or not. 1 

Hydrolysis of purified straw cellulose with 72 per cent sulphuric 
acid gave results in close agreement with those for cotton, thus 
affording evidence that the two celluloses are identical in consti- 
tution. 2 Wood cellulose is also identical with cotton in that it 
gives comparable yields of cellulose triacetate, a-methylglucoside, 
and crystalline dextrose. 3 There is no difficulty in agreeing with 
these conclusions, based on the purified cellulose, but the relation 
of the original hexosans and pentosans to the residual cellulose is 
still unexplained. 

As might be expected, pentosans are removed with greater 
difficulty from lignocellulose than from the isolated cellulose. 
Fromherz 4 purified aspen wood with 5 per cent hydrochloric acid, 
water, alcohol, ether, and 5 per cent ammonia. It was then 
extracted six times with 10 per cent sodium hydroxide for 36 
hours each. In this way about 18 per cent of wood gum was 
obtained, though the wood lost 45 per cent in weight by the 
treatment. The residual wood gave 2.29 per cent furfural and 
0.5 per cent "methylfurfural." 

Extraction of pentosans with dilute sodium hydroxide rapidly 
reaches a maximum when the wood is simultaneously ground 
to a gelatinous condition. 5 When the wood was ground with an 
added solution of pentosans, no adsorption of pentosans took 

1 E. Heuser and W. Dammel, Cellulosechemie, 5 (1924), 45. 

2 E. Heuser and A. Haug, Z. angew. Chem., 31 (1918), 172. 

3 E. Heuser and S. S. Aiyar, Z. angew. Chem., 37 (1924), 27. 
* K. Fromherz, Z. physiol. Chem., 50 (1906), 209. 

5 A. W. Schorger, /. Ind. Eng. Chem., 16 (1924), 141. 



172 CHEMISTRY OF CELLULOSE AND WOOD 

place, though its gelatinous condition presumably should have 
favored this phenomenon. 

It is conceivable that if the carbohydrates are adsorbed, their 
colloidal state might render extraction with alkalis slow and 
incomplete. This argument cannot, however, explain the diffi- 
culty encountered in removing pentosans by acid hydrolysis. 
Winterstein 1 heated beech wood, containing 26.46 per cent of 
xylan, with 1.25 per cent sulphuric acid for 3 hours. The wood 
lost 17.75 per cent in weight but the residue contained 18.46 per 
cent of pentosans. After 3 hours' heating with 5 per cent acid, 
it still contained 10.16 per cent of xylan. Beech-wood cellulose 
isolated with F. Schulze 's reagent retained 21.83 per cent of the 
xylan originally present. A second treatment did not lower the 
xylan content. Similar results were obtained by Schulze. 2 
Winterstein concluded that xylan is present in wood in two 
different forms, the one being destroyed by boiling, dilute acid, 
and F. Schulze's reagent, while the other is resistant; and that 
some of the xylan is chemically combined with the cellulose. 

A similar conclusion had been reached by Schulze and Tollens, 3 
as a result of the difficulty experienced in removing pentosans 
from spent malt. The latter, after boiling with 4 per cent sul- 
phuric acid for 6 hours, followed by extraction with 5 per cent 
sodium hydroxide, still contained pentosans. 

Wheat-bran cellulose yields furfural even after being submitted 
to drastic chemical reactions. Sherman, 4 therefore, concludes 
that about 10 per cent of pentosan is combined with the cellulose. 

According to Kunz 5 barley chaff contains two kinds of 
pentosans. The one is easily hydrolyzed by heating for several 
hours with 2 per cent sulphuric acid; the other with difficulty, 
heating with 1 per cent acid at 3 to 4 atmospheres' pressure being 
required. 

Nitrated wood and straw celluloses of high pentosan content 
contain considerable amounts of unnitrated material. This 
condition is attributed to formation by the pentosans of a hard, 

1 E. Winterstein, Z. physiol. Chem., 17 (1893), 381. 

2 E. Schulze, Z. physiol. Chem., 16 (1892), 435. 

3 E. Schulze and B. Tollens, Ann., 271 (1892), 55. 

4 H. C. Sherman, J. Am. Chem. Soc, 19 (1897), 300. 

5 E. Kunz, Biochem. Z., 74 (1916), 312. 



HEMICELLULOSES AND WOOD CELLULOSE 173 

protective coating around the fibers. 1 Local concentration of 
pentosans would be very strong proof against chemical 
combination, but such a condition has not been established. 

Evidence has been introduced that mannocelluloses exist. 
Schulze 2 called coffee-bean cellulose a mannocellulose, since after 
boiling with dilute hydrochloric acid, followed by treatment with 
F. Schulze's reagent, (20 parts of nitric acid, sp. gr. 1.16, and 3 
parts potassium chlorate), it still contained mannose. Manno- 
cellulose from coffee beans, sesame cake, and coconut cake is not 
destroyed by F. Schulze's reagent, Hoffmeister's reagent (HC1 + 
KC10 3 ), Lifschutz's mixture of sulphuric and nitric acids, nor by 
heating with caustic alkali at 180°; hence it should be considered 
as similar to cellulose. 3 

According to Gilson, 4 the mannocellulose of Schulze does not 
exist as a chemical entity but as a mixture that can be separated 
into mannan and cellulose. Coffee beans, purified with dilute 
acids and alkalis, were dissolved in cuprammonium solution. 
On treating this solution with .carbon dioxide for 15 minutes, 
the cellulose formed a gelatinous precipitate, while mannan 
remained in solution. If mannan exists as such, it is difficult to 
understand how it resists the drastic chemical treatment to which 
it was submitted by Schulze unless the carbohydrate mixture 
was in such coarse particles that the mannan was very slowly 
attacked. 

The status of the mannan in wood is less doubtful. Bertrand, 5 
for no apparent reason, considered the mannose of the conifers 
to exist as a mannocellulose. This view is supported by Sher- 
rard, 6 since mannose persists in sulphite pulp, Cross and Bevan 
cellulose, and the a-, /?-, and 7-celluloses. Mannose was not 
obtained from cotton on hydrolysis, showing that mannose 
residues must preexist in wood. Schorger 7 had found that man- 
nose could be quantitatively hydrolyzed from wood by the aid 

1 B. Rassow and E. Dorr, J. prakl. Chem., 108 (1924), 113. 

2 E. Schulze, Z. physiol. Chem., 16 (1892), 425. 

3 E. Schulze, Z. physiol Chem., 19 (1894), 38. 

4 E. Gilson, La Cellule, 9 (1893), 428. 

5 G. Bertrand, Compi. rend., 129 (1899), 1025. 

6 E. C. Sherrard and G. W. Blanco, J. Ind. Eng. Chem., 15 (1923) 615; 
E. C. Sherrard and A. W. Froehlke, J. Am. Chem. Soc, 45 (1923), 1729, 

7 A. W. Schorger, J. Ind, Eng. Chem., 9 (1917), 748. 



174 CHEMISTRY OF CELLULOSE AND WOOD 

of boiling dilute mineral acids; therefore, unless Schulze's defini- 
tion be discarded, the mother substance of the mannose must be 
classed as a hemicellulose. Heuser 1 does not consider mannan 
as chemically combined with cellulose in sulphite pulp, since the 
greater portion can be removed by extracting the pulp two or 
three times with 17 per cent sodium hydroxide. 

The very interesting observation has recently been made by 
Sherrard 2 that white-spruce cellulose, as ordinarily prepared by 
the Cross and Bevan method, may contain as much as 20 per cent 
of a material extractable by boiling water. It may be isolated in 
the form of a white, extremely hygroscopic powder, having a 
relatively low reducing value. The powder on hydrolysis gives 
about 30 per cent of mannose and 20 per cent of pentose. The 
compound has been obtained in a partially crystalline state, but 
whether it consists of a definite body or a mixture remains to be 
determined. The mannose removed by water extraction consti- 
tutes about 43 per cent of that contained in the original cellulose. 

It is apparent that the relation of certain hexosans and 
pentosans to the " normal cellulose" in wood is still only a 
matter of individual opinion. No convincing arguments or 
experimental data, for or against chemical combination, have 
been produced. 

1 E. Heuser and W. Dammel, Cellulosechemie, 5 (1924), 45. 

2 E. C. Sherrard and G. W. Blanco, J. Ind. Eng. Chem., 15 (1923), 1166. 



CHAPTER VI 
THE CONSTITUTION OF CELLULOSE 

Cellulose can be hydrolyzed to glucose in practically quantita- 
tive yield, showing that it consists entirely of anhydroglucose 
units. Each unit contains three free hydroxyl groups, one of 
which is primary, the remainder secondary. Their position is 
established by the formation of 2, 3, 6-trimethylglucose. Acetol- 
ysis of cellulose gives from 35 to 60 per cent of cellobiose, indicat- 
ing that the glucose units are united wholly or in large part by 
cellobiose linkages. Cellulose -is probably a polycyclic com- 
pound and should be represented as (C 6 Hio0 5 )n, not (C 6 Hi O 5 )n- 
H2O. 1 The number of units in the molecule or aggregate is 
entirely unknown. Cellulose shows a high optical activity in 
cuprammonium solution, but this has no bearing on its structure. 
Rontgen-ray spectrograms of cellulose fibers indicate a crystal- 
line structure. 

Acetolysis of Cellulose. — Acetolysis represents a reaction 
wherein the hydroxyl groups formed by the splitting of ether 
linkages are acetylated 2 and may be considered as simultaneous 
hydrolysis and acetylation. Acetolysis is usually conducted with 
acetic anhydride, or acetyl bromide and chloride in the presence 
of catalysts, such as phosphorus trichloride, zinc chloride, sul- 
phuric acid, and anhydrous sodium acetate. The products 
formed in the acetolysis of cellulose depend on the nature and 
amount of the catalyst and the degree of heating. 

In 1879 Franchimont 3 acetylated cellulose, obtaining a com- 
pound containing 68.5 per cent acetic acid and melting at 212°. 

1 H. Kiliani, Chem. Ztg., 32 (1908), 366; P. Bary, Compt. rend., 178 
(1924), 1159. 

2 E. Knoevenagel, Ann., 402 (1914), 113. 

3 A. P. N. Franchimont, Ber., 12 (1879), 1941-1942. 

175 



176 CHEMISTRY OF CELLULOSE AND WOOD 

Saponification of this derivative gave a new disaccharide at first 
called cellose, 1 then cellobiose, 2 to distinguish it from the monoses. 
Up to this time, no sugar higher than glucose had been obtained 
from cellulose, and the potential importance of cellobiose in 
relation to the constitution of cellulose was immediately recog- 
nized. Acetolysis in the presence of sulphuric acid 2 gives a-cello- 
biose octacetate, and of sodium acetate, 3 /3-cellobiose octacetate. 
The /3-form can be converted into the a-form by heating with 
acetic anhydride in the presence of a small amount of sulphuric 
acid, 4 or preferably zinc chloride. 5 The /3-form can also be 
obtained from acetochloro-cellobiose with silver acetate. 6 The 
production of a-cellobiose octacetate is no indication that it is 
derived from either a- or /?-cellobiose in the cellulose molecule, 
since either form would yield the same «-octacetate. 5 

To prepare cellobiose octacetate, 10 grams of air-dry cotton are 
gradually added to a cooled mixture of 40 grams of acetic anhy- 
dride and 5 grams of concentrated sulphuric acid, care being taken 
that the temperature does not exceed 40 to 50°. Let stand at 
room temperature. Complete solution of the cellulose takes 
place in a few hours and crystallization of the cellobiose octace- 
tate is complete in 6 to 7 days. The crystalline paste is warmed 
with 60 cubic centimeters of glacial acetic acid and poured into 
1500 cubic centimeters of ice water. After standing 30 minutes, 
filter off the octacetate, dry at 100°, and extract with chloroform. 
After evaporation of the solvent the residue is crystallized from 
boiling alcohol. 7 Freudenberg 8 maintains the acetylating solu- 
tion, containing the cellulose, in a cooling mixture for 8 hours, then 
in water at room temperature for 12 hours. It is then removed 
and allowed to stand for 12 days. The yield is 35 to 36 per cent 
of the theoretical in a single operation. 

1 Z. H. Skraup and J. Konig, Ber., 34 (1901), 1115; 32 (1899), 2413. 

2 Z. H. Skraup and J. Konig, Monatsh., 22 (1901), 1011-1036. 

3 L. Maquenne and W. Goodwin, Bull. soc. chim., [3] 31 (1904), 854-859. 

4 L. Maquenne and W. Goodwin, I.e.; E. R. von Hardt-Stremayr, 
Monatsh., 28 (1907), 63-72. 

5 C. S. Hudson and J. M. Johnson, /. Am. Chem. Soc, 37 (1915), 1278. 

6 E. Geinsperger, Monatsh., 26 (1905), 1459. 

7 J. Madsen, "Die Acetolyse der Cellulose zu Cellobiose- und Dextrin- 
acetaten," Diss. Hanover (1917), 53 pp. 

8 K. Freudenberg, Ber., 54 (1921), 771. 



THE CONSTITUTION OF CELLULOSE 177 

The process may be hastened, at the expense of the yield, by 
adding 20 grams of air-dry filter paper, cut into strips, to a mixture 
of 80 cubic centimeters of acetic anhydride and 11 cubic centi- 
meters of concentrated sulphuric acid, the temperature being 
held just below 20°. The flask is then placed in a calcium 
chloride bath held at 120° and as soon as the color of the solution 
changes from red to black, the critical stage, the contents are 
poured into 1.5 liters of cold water. After standing 6 hours the 
precipitate is filtered off, dried at 40°, and crystallized from boil- 
ing 90 per cent alcohol. The yield is 25 to 35 per cent of the 
weight of the cellulose. 1 

Air-dry cellulose containing 8 to 10 per cent of moisture should 

be used. If more water is present, too much heat is generated, 

while perfectly dry cellulose is not penetrated by the acetylating 

mixture. Hydrocellulose, owing to its fineness, generally gives 

better yields than fibrous cellulose. The mixing of the acetic 

anhydride and sulphuric acid should be done with cooling. Sul- 

/COOH 
phoacetic acid, CH 2 \ , is a poorer catalyst than the 

N3O3H 

acetylsulphuric acid, CH3.COO.SO3H, formed at low tempera- 
tures. 2 If the acetylation mixture contains less than 3.5 per cent 
or more than 30 per cent of sulphuric acid, the cellobiose octace- 
tate will not crystallize out. 3 

The yields of cellobiose octacetate are too low to determine if 
cellulose consists entirely of cellobiose residues. Madsen 4 
obtained 32 per cent of cellobiose octacetate directly, this figure 
being increased to 37 to 43 per cent of the theoretical by recovery 
from the dextrin acetates. Ost 5 obtained 37.2 per cent of pure 
cellobiose octacetate and 54.9 per cent of ether-soluble dextrin 
acetates. Comparable yields were obtained by Freudenberg, 6 
who, from a study of the loss of cellobiose by hydrolysis during 
acetolysis, concluded that cellulose contains at least 60 per cent of 
cellobiose. A similar conclusion was reached by Karrer and 

1 W. N. Haworth and E. L. Hirst, J. Chem. Soc, 119 (1921), 197. 

2 W. Schliemann, Ann., 378 (1911), 366-381; Diss. Hanover (1910). 

3 F. Klein, Z. angew. Chem., 25 (1912), 1409-1415. 

4 J. Madsen, I.e. 

5 H. Ost, Ann., 398 (1913), 338. 

6 K. Freudenberg, I.e. 



178 CHEMISTRY OF CELLULOSE AND WOOD 

Widmer. 1 The statement has recently appeared that sulphite 
pulp gives a 90 per cent yield of cellobiose octacetate. 2 

Cellobiose. — Cellobiose forms fine white crystals of the 
composition C12H22O11, after drying at 100°. It decomposes at 
225°. The initial rotation [a] D = +26.1° becomes constant 
at + 33. 7°. 3 The osazone crystallizes in yellow needles melting 
at 208 to 210°. 4 The sugar is scarcely affected by yeast. Cello- 
biase, found in the mycelium of Aspergillus niger and in the 
kernels of apricots and almonds, splits cellobiose into glucose. 5 
Taka-diastase 6 contains a cellobiase that hydrolyzes cellobiose 
completely to glucose in 5 days at a temperature of 37°. Bac- 
teria that ferment cellulose attack cellobiose, but not Ost's 
"celloisobiose." Bacillus butylicus attacks both cellobiose and 
"celloisobiose." 7 

Cellobiose has been synthesized by the action of emulsin on a 
mixture of ethylene glycol, dextrose, and aqueous alcohol. 8 

To prepare cellobiose, dissolve 12 grams of potassium hydroxide 
in 40 cubic centimeters of alcohol at 90°. Cool in a water bath, 
add 10 grams of finely pulverized cellobiose octacetate with 
agitation, taking care that the temperature does not exceed 
35 to 40°. There separates a viscous mass of the potassium salt 
of cellobiose. When this has collected together, decant the 
supernatant liquid, wash the residue twice with strong alcohol, 
dissolve the remainder in a little warm water, and neutralize 
exactly with perchloric acid. Filter to remove the potassium 
perchlorate, concentrate to the point of formation of a thin film, 
let cool, filter a second time, and concentrate again on the water 
bath to the point of crystallization. Decant into a small flask, 
add 3 to 4 volumes of methyl alcohol, and let stand. The follow- 

1 P. Karrer and F. Widmer, Helvetica Chim. Acta, 4 (1921), 174-184. 

2 T. Ozawa, J. Chem. Ind. (Japan), 27 (1924), 884; J. Soc. Chem. Ind., 
44 (1925), B166. 

3 Z. Skraup and J. Konig, Ber., 34 (1901), 117. 

4 W. SCHLIEMANN, l.C, 371. 

5 G. Bertrand and M. Holderer, Compt. rend., 149 (1909), 1385-1387; 
150 (1910), 230-232; E. Fischer and G. Zemplen, Ann., 372 (1909), 254- 
256. 

6 C. Neuberg and O. Rosenthal, Biochem. Z., 143 (1923), 399. 

7 C. Neuberg and R. Cohn, Biochem. Z., 139 (1923), 527. 

8 E. Bourquelot and M. Bridel, Compt. rend., 168 (1919), 1016-1019. 



THE CONSTITUTION OF CELLULOSE 



179 



Derivatives of Cellobiose 



Substance 


Formula 


Melting 
point 


Wd 


Refer- 
ence 




Cl2Hl60 4 (C2H 3 2 )7 

Cl2H 14 03(C2H302)8 
Cl2Hl 4 03(C2H 3 2 )8 

Ci 2 Hi404(OCH 3 )7 
Ci2Hi 3 03(OCH 3 )8 

Cl2Hl 4 03(C2H 3 02)7Cl 

Ci2Hi4C3(C 2 H302)7Br 

Cl2Hl403(C2H 3 02)7l 
CisHib04(C7H:02)7 


195-197° 
229.5 

202 

83-84 

86 

195 

180 + 

160-170 

202-204 


+ 19.95° 
+ 41.95 

- 14.74 

- 7.73 

- 15.91 
+ 75.21 
+ 96.3 
+ 125.3 


4 




\ , 




r l 


Hexamethyl-methylcellobioside 

Heptamethyl-methylcellobioside 


■ 

3 




5 











i C. S. Hudson and J. M. Johnson, J. Am. Chem. Soc, 37 (1915), 1278. 
2 P. Karrer and F. Widmer, Helvetica Chim. Acta, 4 (1921), 182. 
3Z. H. Skraup and E. Geinsperger, Monatsh., 26 (1905), 1469. 
«E. Fischer and G. Zemplen, Ber., 43 (1910), 2536. 
5 S. V. Hintikka, Cellulosechemie, 4 (1923), 62. 

ing day collect the cellobiose that has deposited and wash care- 
fully with methyl alcohol. The yield is 3 grams. 1 

Hess, 2 by the action of acetyl chloride on cellulose, and subse- 
quent saponification, obtained an anhydrobiose, C12H20O10— 
2H 2 0, becoming brown at 250° and decomposing at 270°. The 
sugar was insoluble in water and in ammonia, soluble in the 
caustic alkalis and in cuprammonium, and reduced Fehling's 
solution only slightly. It had a marked tendency to association. 
It was considered as closely related to the structural unit in 
cellulose. Acetolysis of ethylcellulose 3 gave a tetraethylanhydro- 
biose, as shown by the molecular weight which was determined in 
very dilute aqueous solution to prevent association. 

The sulpholysis of cellulose with fuming sulphuric acid at 0° 
gave a sulphate which is apparently a derivative of cellobiose. 4 
The crystalline barium salt, (Ci 2 H2iOio-S0 4 )Ba, which is obtained 
in a yield of 50 per cent of the theoretical, is soluble in water, but 
insoluble in alcohol. It is d-rotatory, and reduces Fehling's 
solution only on warming. 



1 L. Maquenne and W. Goodwin, Bull. soc. chim., [3] 31 (1904), 855. 
3 X. Hess, Ber., 54 (1921), 2867-2885. 

3 K. Hess and W. Wittelsbach, Ber., 54 (1921), 3232-3241; cf. K. Hess 
et al., Z. angew. Chem., 34 (1921), 449-454. 

4 A. C. von Euler and K. Melander: cited by E. Hagglund, "Die 
Hydrolyse der Zellulose und des Hopes' 7 (1915), p. 15. 



180 CHEMISTRY OF CELLULOSE AND WOOD 

Procellose. — The acetolysis of cellulose, using sulphuric acid as 
the catalyst, yields also the acetyl derivative of a trisaccharide, 
procellose. 1 It crystallizes with two molecules of water, melts 
at 210°, and has the specific rotation [a] D = +22.8°. It con- 
tains one aldehyde group . The following constitution is suggested : 



CH 2 OHCHOHCH(CHOH) 2 CH 



i 



CH 2 OHCHCH (CHOH) 2 CH 

A 

CH 2 OHCH.(CHOH) 3 CHO 

The "celloisobiose" of Ost and Prosiegel 2 has been shown to be a 
mixture of cellobiose and procellose. 3 When the mixture of 
sugars is shaken for several days at room temperature with 70 
per cent alcohol, with daily renewal of the solvent until the 
amount of sugar dissolved in the solvent becomes constant, the 
residue consists of procellose which is purified by a recrystalliza- 
tion from 95 per cent alcohol. 

Constitution of Cellobiose and Cellulose. — On treating cotton 
impregnated with 17 per cent sodium hydroxide, in quantity 
represented by C 6 Hio05.2NaOH, with dimethylsulphate, Den- 
ham and Woodhouse 4 obtained a product with 24 to 26 per cent 
methoxyl. Hydrolysis of this methylcellulose by Willstatter's 
method gave a trimethylglucose (m.p.110 ; [a] D = +69.52°), 
in addition to lower methylglucoses. The trimethylglucose 
behaved as a uniform body, indicating that the cellulose molecule 
was symmetrical and contained but three free hydroxyl groups 
for each C 6 residue. There are four possible trimethylglucoses, 
each of which can exist in a- and /3-form. 

1 G. Bertrand and S. Benoist, Compt rend., 176 (1923), 1583-1587. 

2 H. Ost and R. Prosiegel, Z. angew. Chem., 33, 1 (1920), 100; H. Ost and 
G. Knoth, Cellulosechemie, 3 (1922), 25. 

3 G. Bertrand and S. Benoist, Compt. rend., 177, (1923) 85-87. 

4 W. S. Denham and H. Woodhouse, J. Chem. Soc, 105 (1914), 2357- 
2368; 103 (1913), 1735-1742. 



THE CONSTITUTION OF CELLULOSE 



181 



I 
CHOH 



II 
-CHOH 



III 
CHOH 



IV 
CHOH 



CHOH / CHOCH3 / CHOCH3 / CHOCH3 

I o I I o 

CHOCH3 \ CHOCH3 \ CHOCH3 \ CI 



CR 



CHOCHc 



CH 



CHOCH s 




CH 2 OH 




CHOCHr 



CHoOCH, 



CH 2 OH 



CH 2 OCHj 



CHoOCH. 



r i20 o 



= +42.7° 



I. 3, 5, 6-Trimethylglucose. 

II. 2, 3, 5-Trimethylglucose ; syrup; 
66.8°; n D l. 4780. 1 

HI. 2, 3, 6-Trimethylglucose; fine needles or short prisms; 
m.p. 122 to 123° or 92 to 93°; [a] D = +90.2° -> 70.5°; n D = 
1.4743. 2 

IV. 2, 5, 6- Trimethylglucose. 

For establishing the identity of the trimethyl glucose from 
cellulose, formula (I) is excluded since it can give an osazone, 
and (II) is known. The trimethylglucose from cellulose on 
condensation with hydrocyanic acid gave an acid containing only 
two methoyxl groups. Titration showed that the acid was 
partially lactonized. On the assumption that a 7-lactone is 
formed, the methyl group at position 3 must have been split off 
so that the trimethylglucose would have structure III. 3 The 
correctness of this formula has been confirmed by Irvine and 
Hirst. 4 

By repeated methylation it was possible to obtain methyl- 
cellulose containing 44.6 per cent methoxyl, 6 the theoretical being 
45.6 per cent. Products containing 40 per cent or more of 

1 J. Irvine and J. W. Oldham, /. Chem. Soc, 119 (1921), 1756; cf. T. 
Purdie and R. C. Bridgett, Ibid., 83 (1903), 1037. 

2 J. C. Irvine and E. L. Hirst, /. Chem. Soc, 121 (1922), 1216. 

3 W. S. Denham and H. Woodhouse, J". Chem. Soc, 111 (1917), 244-249. 

4 J. C. Irvine and E. L. Hirst, /. Chem. Soc, 121 (1922), 1217. 

5 W. S. Denham, /. Chem. Soc, 119 (1921), 77-81. 



182 



CHEMISTRY OF CELLULOSE AND WOOD 



methoxyl are soluble in water but insoluble in cuprammonium 
solution. 

The constitution proposed for cellobiose by Haworth and 
Leitch 1 has been substantiated by the hydrolysis of heptamethyl- 
0-methylcellobioside to 2, 3, 5, 6-tetramethylglucose, and 2, 3, 6- 
trimethylglucose. 2 The trimethylglucose (m.p. 115 to 116°; 
[<*]d = +68.7°) was identical with that prepared from methyl- 
ated lactose 3 and methylated cellulose. 4 Hudson 5 had shown 
from the optical rotations of lactose and cellobiose that these 
sugars must contain the same glucose residue. Cellobiose is, 
accordingly, a glucose-5-/3-glucoside. 

CH 2 OH 




CH 2 OH 



Cellobiose 



ii 



CH- 



o 



CH0CH3 

CHOCH 3 

u 

CHOCH3 
CH20CH3 



CH20CH3 
CJH 

in 



CHOCHr 

CHOCH3 

!HOCH 3 



+2H2O O 



/CHOH 
CHOCH3 
CHOCH3 

^CH 
CHOCH3 
CH2OCH3 



III 



CH2OCH3 



CHOH 

CH +CH3OH 

CHOCH3 

CHOCHs 

:hoh 



1 W. N. Haworth and G. C. Leitch, J. Chem. Soc, 115 (1919), 813. 

2 W. N. Haworth and E. L. Hirst, J. Chem. Soc, 119 (1921), 193; 
P. Karrer and F. Widmer, Helvetica Chim. Acta, 4 (1921), 295; cf. M. 
Bergmann, Naturwiss., 9 (1921), 308. 

3 W. N. Haworth and G. C. Leitch, J. Chem. Soc, 113 (1918), 188. 

4 W. S. Denham and H. Woodhouse, J. Chem. Soc, 105 (1914), 2364. 

5 C. S. Hudson, J. Am. Chem. Soc, 38 (1916), 1573. 



THE CONSTITUTION OF CELLULOSE 183 

1 . Heptamethyl-/3-methylcellobioside . 
II. 2, 3, 5, 6-Tetramethylglucose. 
III. 2, 3, 6-Trimethylglucose. 

The formation of 2, 3, 6-trimethyglucose led Haworth and 
Hirst 1 to suggest for cellulose the structure given below, in which 
X represents additional glucose anhydride groups. The same 
structure was simultaneously proposed by Hibbert. 2 

CH 2 OH 

CH— O— X 

CH 

CHOH 
O | 

CHOH 

I 
CH— O— X 

Cellulose 
(Unit group) 

It had been shown by Monier- Williams 3 that hydrolysis of 
cellulose with 72 per cent sulphuric acid yielded 91 per cent of 
crystalline glucose. Furthermore, acetolysis of cellulose, followed 
by simultaneous hydrolysis and formation of methylglucoside 
through the heating of the acetylated products with methyl 
alcohol and hydrogen chloride, gave 85 per cent of the theoretical 
yield of methylglucoside. 4 On repeating the acetolysis with 
sulphuryl chloride as the catalyst, a practically quantitative yield 
of pure cellulose triacetate was obtained. The triacetate was 
converted into an equilibrium mixture of a- and /?-methylgluco- 
sides in a yield of 95 per cent of the theoretical. 5 This showed 
that cotton is composed entirely of anhydroglucose residues. 

1 W. N. Haworth and E. L. Hirst, J. Chem. Soc, 119 (1921), 196. 

2 H. Hibbert, J. Ind. Eng. Chem., 13 (1921), 256, 334. 

3 G. W. Monier-Williams, /. Chem. Soc., 119 (1921), 803. 

4 J. C. Irvine and C. W. Soutar, /. Chem. Soc., 117 (1920), 1489-1500. 

5 J. C. Irvine and E. L. Hirst, J. Chem. Soc, 121 (1922), 1585-1591. 



184 



CHEMISTRY OF CELLULOSE AND WOOD 



It was now known that cellulose on degradation gave some 
cellobiose, and a nearly quantitative yield of glucose. Further- 
more, the formation of 2, 3, 6-trimethylglucose showed that in a 
portion of the cellulose molecule positions 1 and 5 were protected 
from methylation, but whether all the glucose residues were alike 
remained unknown. This point was clarified by Irvine and 
Hirst 1 in a continuation of their neat research. A trimethyl- 
cellulose, which contained 43 per cent methoxyl after 20 methyla- 
tions, was heated with methyl alcohol and hydrogen chloride. 
There was obtained 91.5 per cent of the theoretical yield of an 
equilibrium mixture of a- and /?- 2, 3, 6-trimethylglucosides. The 
mixture on hydrolysis gave 86 per cent of the theoretical yield 
of 2,3, 6-trimethylglucose. No isomeric trimethylglucose was 
present. This proved that all the glucose residues in cellulose 
have the hydroxyl groups in positions 2, 3, and 6 unsubstituted. 
Since the highest yield of cellobiose claimed did not exceed 60 
per cent, cellulose as a matter of precaution was given the struc- 
ture of a symmetrical tri-1, 5-anhydroglucose. 



O 



CH 2 OH ^^O 
,CH— O— CH.CH.CH(OH).CH(OH).CJi 

CHOH 

I 

CHOH O 

J 
'CH 

CH— O— CH.CH(OH).CH(OH).CH.CH.CH 2 OH 



CH 2 OH 



Cellulose (Irvine) 



i J. C. Irvine and E. L. Hirst, J. Chem. Soc, 123 (1923), 518-532; J. Soc. 
Chem. Ind., 41 (1922), 362-363R; cf. Ibid., 44 (1925), 242. 



THE CONSTITUTION OF CELLULOSE 



185 



It is now possible to exclude most of the other structures pro- 
posed for cellulose, owing to their inability to furnish quantitative 
yields of 2, 3, 6-trimethylglucose. 1 

It has been shown in the chapter on the distillation of cellulose 
that the formation and structure of /3-glucosan has little bearing 
on the constitution of cellulose. On treatment with chloral and 
sulphuric acid, a-glucosan gives parachloralose and /3-glucosan 
gives chloralose. 2 Cellulose under the same treatment gives 
dichloralglucoses. 3 

Karrer* considered both starch and cellulose to be built up of 
C12H20O10 units. Cellulose was looked upon as a dimeride of 
anhydrocellobiose, cellosan, polymerization taking place through 
subsidiary valencies without rupture of the oxygen bridges, the 
arrangement and cohesive forces being similar to those existing 
in a crystal. 



CH 2 OH.CH.CH.CH(OH).CH(OH).CH 



O 







CH.CH(OH).CH(OH).CH.CH.CH 2 OH 



Cellulose (Karrer) I 

This formula was subsequently changed to (II), 5 since triacetyl- 
cellulose on heating with phosphorus pentabromide gave tri- 

iL. Vignon, Compt. rend., 127 (1898), 873; C. F. Cross and E. J. Bevan, 
/. Chem. Soc, 79 (1901), 366; A. G. Green and A. G. Perkin, J. Chem. Soc, 
89 (1906), 811; A. G. Green, Z. Farben-u. Textil. Chem., 3 (1904), 97; H. 
de Mosenthal, /. Soc. Chem. Ind., 30 (1911), 782; K. Gebhard, Chem. 
Ztg., 37 (1913), 663; B. Tollens, "Kohlenhydrate," Leipzig (1914), p. 564; 
H. Barthelemy, Caoutchouc & gutta-percha, 14 (1917), 9274; K. Hess, 
Z. Elektrochem., 26 (1920), 232; Helvetica Chim. Acta, 3 (1920), 620; A. C. 
von Euler, Chem. Ztg., 45 (1921), 977. 

2 A. Pictet and F. H. Reichel, Helvetica Chim. Acta, 6 (1923), 621-627. 

3 J. H. Ross and J. M. Payne, J. Am. Chem. Soc, 45 (1923), 2363-2366. 

4 P. Karrer, Cellulosechemie, 2 (1921), 127. 

5 P. Karrer and A. P. Smirnov, Helvetica Chim. Acta, 5 (1922), 187-201. 



186 



CHEMISTRY OF CELLULOSE AND WOOD 



6 5 4 3 2 1 

CH 2 .CH(OH).CH.CH(OH).CH(OH).CH 



aO 



CH.CH(OH).CH(OH).CH 
12 3 4 



06 



CH.CH 2 0H 
5 6 



Cellulose (Karrer) II 



aceto-l,6-dibromoglucose (m.p. 175°; [a]i 9 ° = +192°). It was 
shown that with pentacetylglucose the acetyl group in position 
6 was not replaced by bromine under similar treatment, and 
that acetodibromoglucose was not obtained from cellobiose 
octacetate. Fission of the ether linkage at a would give cello- 
biose and at b maltose. Since the upper half of the formula would 
give 2, 3, 5-trimethylglucose, the structure is incompatible with 
Irvine's findings that cellulose gives only 2, 3, 6-trimethylglucose. 
The formation of triaceto-l,6-dibromoglucose is, however, dis- 
concerting unless it is assumed that there has been a migration of 
the acetyl group from position 6 and a replacement by bromine. 



CH O- 



CH< 



/ CHOH 



O 



CHOH 



CH 



CHOH 




CHOH 

I 
CH 

CHOH 

I 
CHOH 



CH 2 OH \CHOH 

Maltose 1 

This reaction would be expected from Fischer and Armstrong's 2 
preparation of triaceto-l,6-dibromoglucose by the action of 



1 W. N. Ha worth and G. C. Leitch, /. Chem. Soc, 115 (1919), 809. 

2 E. Fischer and E. F. Armstrong, Ber., 35 (1902), 833. 



THE CONSTITUTION OF CELLULOSE 



187 



hydrogen bromide on glucose pentacetate. Zechmeister 1 has 
shown that the presence of hydrogen bromide is essential to the 
action of acetyl bromide on cellulose and as degradation proceeds 
there is a partial replacement of acetyl groups by bromine. 

A formula has been suggested by Schorger 2 in which four ortho- 
glucosan groups are symmetrically arranged to form a parallelo- 
piped. This poly cyclic structure explains the resistance of 



I CH2OH 

O— CH 

(7) 



CJH- 



CHOH 
CHOH 

A— 



-CH 

CHOH 
CHOH 

c!h- 



(3) 
-O- 



CH 2 OH 
I 
-CH 

I 

CHOH 



(0) 
JO' 



i 



H20H 



(4) 



CHOH 
CH-^(5) 



-CH 

CHOH 
CHOH 

¥— 

CH2OH 



(8) 
-O 



cellulose to hydrolysis, the quantitative formation of 2,3,6- 

trimethylglucose, and the possibility of obtaining a quantitative 

yield of either cellobiose or an isomer, isocellobiose. The 

customary small yields of cellobiose are due to equal chances for 

the formation of the two disaccharides, the isomer being the more 

readily hydrolyzed. Fission at (1), (2), (4), (5), and (6) would 

give cellobiose, and at (7), (8), (1), (3), and (4) isocellobiose. 

Acceptance of this formula depends on the isolation of a disac- 

charide having the constitution of isocellobiose. 

CH 2 OH 

I 
CHOH 



CH 



CHO 



O 



CHOH CHOH 

I I 
CHOH CHOH 

I I 
CH— O— CH 

I 
CHOH 



CH 2 OH 
Isocellobiose 

1 L. Zechmeister, Ber., 66 (1923), 573-578. 

2 A. W. Schorger, J. Ind. Eng. Chem., 16 (1924), 1274. 



188 CHEMISTRY OF CELLULOSE AND WOOD 

Some interesting experiments have been made by Hibbert 1 and 
his pupils leading to the synthesis of cellulose and related carbo- 
hydrates. Condensation of bromoacetaldehyde with various 
polyatomic alcohols yielded acetals related to the carbohydrates. 
The hydroxyl groups in /,6-positions with reference to the 
carbonyl group appear to be key positions in the condensation of 
monoses to polysaccharides. When y,5-dihydroxybutylmethyl- 
ketone is heated with a trace of mineral acid, it undergoes a cyclic 
acetal condensation to give a resin which is presumably a tri- 
meride. Inulin may be formed from fructose in an analogous 
way. 2 

The polysaccharides are probably heterocyclic compounds, 
and information is greatly needed on the stability of the higher- 
member ed rings. A study 3 of the formation and stability of 
cyclic acetals has shown that six-membered rings are more stable 
and more easily formed than the five or seven. The closed spiral 
ring structure proposed to account for the polymerization of 
cellulose lacks symmetry, and the bridge linkage would appear to 
be particularly susceptible to fission. 

Aside from the probable mode of linkage nothing is known of 
the order of arrangement and number of orthoglucosan groups 
forming the cellulose molecule. The number was formerly con- 
sidered to be very high, the molecular weight having been deter- 
mined by ordinary methods which are not applicable to colloids. 
Geinsperger, 4 from a determination of the chlorine content of 
acetochlorocellulose, arrived at a molecular weight of 5508, which 
is equivalent to (C 6 Hi O 5 )34. Joyner 5 in studying the viscosity 
of cellulose solutions found that 16 grams of oxygen reacted with 
about 2500 grams of cellulose, so that the molecular weight of 
cellulose must be at least 2500 if one atom of oxygen reacts with 
one molecule of cellulose. By the boiling-point method, cellulose 
acetate 6 showed a molecular weight of 6480, which is equivalent 

1 H. Hibbert and H. S. Hill, J. Am. Chem. Soc, 45 (1923), 734-751. 

2 H. Hibbert and J. A. Timm, /. Am. Chem. Soc, 45 (1923), 2433-2439. 

3 H. S. Hill and H. Hibbert, J. Am. Chem. Soc., 45 (1923), 3117-3124; 
3124-3132; cf. Ibid., 46 (1924), 1281, 1283. 

4 E. Geinsperger, Monatsh., 26 (1905), 1467. 

5 R. A. Joyner, /. Chem. Soc, 121 (1922), 2395. 

6 A. Nastjukoff, Ber., 33 (1900), 2241. 



THE CONSTITUTION OF CELLULOSE 189 

to (C 6 Hio0 5 )4o, and nitrocellulose, 1 (C 6 Hi O 5 )i2. High variable 
molecular weights were obtained by Hess, 2 using various cellulose 
derivatives. Trimethylcellulose A, e.g., in aqueous solution, gave 
molecular weights varying from 1400 to 8700. 

Our usual conception of a molecule has undergone great change 
through Werner's 3 theory of primary and secondary valencies, 
and the use of X-ray spectrograms. According to Langmuir, 4 
any solid or liquid body must be considered as a single large 
molecule consisting of atoms held together by chemical forces. 
Cohesion, adsorption, and such changes of state as evapora- 
tion, condensation, solution, and crystallization are chemical 
phenomena. Harrison 5 has suggested that the entire cellulose 
fiber may be regarded as the molecule. 

The use of the terms "polymerization" and " depolymeriza- 
tion" in connection with cellulose should be discontinued, as they 
have a meaning in organic chemistry not applicable to colloids. 
Harries 6 has suggested that they be replaced by aggregation and 
disaggregation. The supposed polymerization of drying oils has 
been shown to be colloidal gelation. Molecular-weight deter- 
minations show that only a trace of condensation has taken 
place in the thickened oil. 7 The term " association" 8 is objec- 
tionable, owing to its established use to describe the weak union 
of two or more molecules of a substance, as shown by molecular- 
weight determinations, by mere solution in a solvent. 

Our present conception of the size of the cellulose molecule 
varies from a large indefinite number 9 of glucose residues held 

1 G. Bumcke and R. Wolffenstein, Ber., 32 (1899), 2495, 2506. 

2 K. Hess, Ann., 435 (1924), 1-144. 

3 A. Werner, "Neure Anschauungen auf dem Gebiete der anorganischen 
Chemie," Braunschweig (1905); F. Henrich, "Theories of Organic Chem- 
istry," N. Y. (1922). 

4 I. Langmuir, J. Am. Chem. Soc, 38 (1916), 2221-2295. 

5 W. Harrison, Second Rept. Colloid Chem. Brit. Assn. Adv. Science 
(1921), p. 55. 

6 C. Harries, Ber., 56 (1923), 1048. 

7 H. Wolff, Z. angew. Chem., 37 (1924), 729; R. S. Morrell, J. Soc. 
Chem. Ind., 43 (1924), 362T. 

8 H. Hess, I.e., 104. 

9 H. Hibbert, /. Ind. Eng. Chem., 13 (1921), 256; J. C. Irvine and 
E. L. Hirst, J. Chem. Soc, 123 (1923), 525; A. C. von Euler, Chem. Ztg., 
45 (1921), 977, 998. 



190 CHEMISTRY OF CELLULOSE AND WOOD 

together by cellobiose linkages to a small definite number. 
Wise 1 pictures cellulose as an aggregate of a large number of 
small units held together by secondary valencies, no suggestion 
being made regarding the size of these units. Zwickker 2 repre- 
sents the molecule as composed of nine glucose residues arranged 
to form the edges of a triangular prism. According to Karrer, 3 
cellulose is a dimeride of anhydrocellobiose units held together 
by subsidiary valencies. 

Hess 4 holds the extreme view that cellulose consists of C 6 Hi O 5 
units held together by association. Through a study of the 
optical rotation of cuprammonium solutions of cellulose of vary- 
ing composition, it was found that these solutions obeyed the law 
of mass action, a condition explainable only on the assumption 
that the chemical unit in cellulose is C 6 Hi O 5 . All forms of 
cellulose free from impurities reached almost instantly a maxi- 
mum rotation at the point corresponding to 1 Cu:l C 6 Hio0 5 . 
This indicates that no ether linkages were ruptured, since the 
hydrolysis of carbohydrates is a time reaction. Saponification of 
cellulose acetate A gave cellulose A that also behaved like pure 
cotton cellulose in cuprammonium solution. It would seem 
hazardous to draw such sweeping conclusions from the behavior 
of a reagent having so special an action on carbohydrates as 
cuprammonium. Grossmann 5 found that in alkaline copper 
solutions the optical rotation of sugars was greatly changed and 
frequently reversed; thus for dextrose [a]hb = +78° became 
-375°, while fructose showed + 1423°. 

Hess was also able to synthesize a derivative of cellobiose in 
the following way: Treatment of cellulose acetate A with a 
mixture of acetyl bromide and hydrogen bromide at — 15° gave 
a new tetracetylbromoglucose, m.p. 60 to 70°; [«]z> 4 ° = ca. + 
80°. On allowing an ether-chloroform solution of this tetracetyl- 
bromoglucose to stand in moist air at 2 to 3° for several days 
there was obtained celloglucosan, C 6 Hio05.H 2 0, m.p. 107 to 109°; 
[a\ D = +89.3° (in water). The tribenzoate of celloglucosan on 

1 L. E. Wise, /. Ind. Eng. Chem., 15 (1923), 711. 

2 J. J. Zwickker, Rec. trav. chim., 41 (1922), 49. 

3 P. Karrer, Cellulosechemie, 2 (1921), 127. 

4 K. Hess et al., Ann., 435 (1923), 1-144; Z. angew. Chem., 34 (1924), 993. 
6 H. Grossmann, Z. Ver. deut., Ziicker-Ind., 56 (1906), 1024. 



THE CONSTITUTION OF CELLULOSE 191 

treatment with acetyl bromide and hydrogen bromide gave 
heptacetylbromocellobiose, which was obtained directly from 
cellulose acetate A by the same treatment. Hess concludes that 
the anhydrocellobiose group does not necessarily exist in the 
cellulose molecule but may be formed by a secondary reaction. 
This possibility is rather remote in spite of the fact that isomal- 
tose 1 and apparently a trisaccharide 2 have been obtained by the 
action of hydrochloric acid on glucose. The bacterial degrada- 
tion of cellulose through cellobiose, 3 under conditions precluding 
the likelihood of enzymatic synthesis, is strong evidence that the 
cellobiose linkage preexists in cellulose. The resistance of cellu- 
lose to hydrolysis renders it difficult to accept a C 3 Hi O 5 entity for 
cellulose, since all the anhydroglucoses so far obtained are readily 
hydrolyzable. To accept Hess' view, it would be necessary to 
assume from available experimental evidence that the secondary 
valencies in cellulose are stronger then the primary valencies, 
which is turning the tables. 

Some interesting results on the structure of cellulose have been 
obtained by the use of X-rays, but the data are too meager to be 
accepted with finality. All methods depend on the diffraction 
of the X-ray waves by the planes of atoms, which have a space 
lattice within the crystal. Only crystalline substances diffract 
X-rays. The original method of Laue depends on passing a 
pencil of X-rays through a thin oriented section of a crystal, a 
point diagram being obtained. In the spectroscopic method 
developed by the Braggs the X-rays are reflected from atomic 
planes parallel to the smooth surface of an oriented crystal. The 
method of most general application consists in subjecting a 
narrow tube, containing the finely powdered substance to be 
examined, to a ribbon-like beam of X-rays. With a finely pow- 
dered crystalline substance, the atomic planes are oriented in all 
directions, but only those planes that chance to lie in the proper 
direction are effective. The diffracted rays are registered on a 
film as a series of parallel lines of varying intensity and character- 
istic pattern (Fig. 4). 



1 E. Fischer, Ber., 23 (1890), 3687. 

2 E. Grimaux and L. Lefevre, Compt. rend., 103 (1886), 146. 

3 H. Pringsheim, Z. physiol. Chem., 78 (1912), 266. 



192 



CHEMISTRY OF CELLULOSE AND WOOD 



The X-ray examination of cellulose indicates that it is crystalline 
(Fig. 5). Cotton, ramie, and wood pulp that had been reduced 



FILM 'AND DiFFRACT/ON 'PHOTOGRAPH 



POWDERED MINERAL 
HIGHLY MAGNIFIED 




r IQ0)n:2 



nut 



ML 



ooo) 



UhlEWATED 



BEAM 



Fig. 4. — Idealized representation of the diffraction of a monochromatic beam 
of X-rays from a crystalline powder. Atomic planes are indicated by dots. 
(After Kerr.) 



1,29- 
1.46- 
170- 
2.G3- 

2.58- 

3.40- 



81111 

mat 




«.,^.- 



Reference line of 
ppim<5.py beam 



Fig. 5. — Spectrograms of bundles of ramie fibers (Boehmeria nivea) taken at 
various angles with reference to the long axes of the fibers. Figures in the 
vertical columns represent the interplanar spacings in A.U. (After Sponsler.) 

to a fine state of division by exposure to the vapors of hydrochloric 
acid gave identical patterns; hence their crystalline forms were 



THE CONSTITUTION OF CELLULOSE 193 

identical. 1 The cellulose spectrogram indicates that the crystals 
are arranged symmetrically with respect to the axis, groups of 
four C 6 Hi O 5 units, arranged in the form of aparallelopiped, being 
regularly repeated. 2 Gonell 3 concludes from the structure of the 
individual points in the patterns of various celluloses that more 
than one crystalline substance must be present. The various 
artificial silks, with the exception of cellulose acetate, show a 
crystalline structure. 4 The difference between the diagrams 
appears in the arrangement of the crystallites with respect to the 
axis of the fiber. Irregular arrangement produces a ring diagram 
and a regular arrangement a point diagram. The artificial silks 
give an intermediate pattern. Cotton treated with a mercerizing 
solution of sodium hydroxide and washed without tension gives 
a ring diagram; if washed under tension the fiber shows a point 
diagram. 5 Tunicin gives the same diagram as ramie cellulose; 
lichenin a different one. 6 

Careful investigations on the X-ray diffraction patterns of 
starch and cellulose have been made by Sponsler. 7 His latest 
values for the interplanar spacings in plant fibers are given below. 8 
The conclusion was recently reached by Sponsler 9 that the 

Interplanar Spacings in Plant Fibers 

A.u. A.u. 

6.10 2.62 

5.40 2.58 

3.98 2.17 

3.40 2.03 

3.20 1.11 

1 R. O. Herzog and W. Jancke, Z. Physik., 3 (1920), 196-198; Ber., 53 
(1920), 2162-2164; R. O. Herzog, W. Jancke, and M. Polanyi, Z. Physik., 
3 (1920), 343-348. 

2 R. O. Herzog, Cellulosechemie, 2 (1921), 101-102; R. O. Herzog and 
W. Jancke, Umschau, 25 (1921), 53-54. 

3 H. W. Gonell, Z. Physik., 25 (1924), 118-120. 

4 R. O. Herzog and H. W. Gonell, Kolloid-Z., 35 (1924), 201-202. 

5 R. O. Herzog and G. Londberg, Ber., 57 (1924), 329. 

6 R. O. Herzog and H. W. Gonell, Z. physiol. Chem., 141 (1924), 63-67. 
7 0. L. Sponsler, Am. J. Botany, 9 (1922), 471-492; J. Gen. Physiol, 

5 (1923), 757-776; Ibid., 9 (1925), 221-233; Science, 62 (1925), 547-548. 

8 O. L. Sponsler, Nature, 116 (1925), 243. 

9 Private communication. 



194 CHEMISTRY OF CELLULOSE AND WOOD 

elementary cell has the dimensions 5.40 X 6.10 X 10.30 A.U., 
which is equal in volume to two CeHioOs groups. It appears to 
be questionable whether the diffraction patterns of plant mate- 
rials are not due to the uniform arrangement with which the units 
are deposited during growth, rather than to a structure which is 
reproducible in vitro and crystalline in the usually accepted sense. 

Optical Rotation. — Solutions of cellulose in concentrated sul- 
phuric or hydrochloric acid, 1 and a mixture of zinc chloride and 
hydrochloric acid, 2 are optically inactive until hydrolysis takes 
place. Bechamp 3 found that cellulose dissolved in hydrochloric 
acid of a concentration of 39.8 per cent remained inactive for 6 
hours. Willstatter and Zechmeister, 4 using a stronger acid, 
found optical activity after one hour. A 1 per cent solution of 
viscose appeared to be weakly dextro-rotatory, but difficulty in 
preparing a clear solution made the readings uncertain. 5 

A cuprammonium solution of cellulose in which the copper had 
been displaced with zinc gave a colorless solution inactive towards 
polarized light. 6 Levallois 7 found that a cuprammonium solu- 
tion of cellulose was optically active, a 1 per cent solution showing 
a rotation of approximately —20° in a 200-millimeter tube. Cot- 
ton, linen, hemp, and Girard's hydrocellulose gave the same rota- 
tion under similar conditions. The rotation was not constant, 
but varied with the concentration of the cellulose and the ratio 
of copper to cellulose. The optical activity was attributed by 
Bechamp 8 to the cuprammonium itself, the cellulose merely 
augmenting the rotation as in the case of boric acid and tartaric 
acid. These contentions were refuted by Levallois. 9 

The optical activity of curpammonium solutions of cellulose 
has been confirmed by Hess and Messmer. 10 The ammonia 

1 A. Bechamp, Compt. rend., 42 (1856), 1213. 

2 J. Konig and F. Huhn, " Bestimmung der Zellulose in Holzarten und 
Gespinstfasern" (1912), p. 54. 

3 A. Bechamp, Compt. rend., 100 (1885), 370. 

4 R. Willstatter and L. Zechmeister, Ber., 46 (1913), 2404. 

5 L. Vignon, Bull. soc. chim., [3J 31 (1904), 107. 

6 E. Mulder, Jahresb. Chemie, (1863), 566. 

7 A. Levallois, Compt. rend., 98 (1884), 732-735. 

■ A. Bechamp, Compt. rend., 99 (1884), 1027; 100 (1885), 279, 368. 
» A. Levallois, Compt. rend., 99 (1884), 1122; 100 (1885), 456. 
« K. Hess, E. Messmer, and E. Jagla, Ber., 55 (1922), 2432. 



THE CONSTITUTION OF CELLULOSE 



195 



remaining constant, the optical activity decreased with the 
decrease in concentration of both copper and cellulose. The 
specific rotation, [a] = ca. —950° to — 1000°, is in agreement with 
the value found by Levallois. Ethylcellulose dissolved in esters, 
viscose, and gelatinized cellulose in neutral salt solutions were 
optically inactive. 1 Cellulose represents a case of latent asym- 
metry, the cuprammonium behaving like boric acid in the case of 




NH 3 
CH, 



mannitol. The cuprammonium forms a complex with the cellu- 
lose through a diversion of the subsidiary valencies holding the 
cellulose groups together. This complex, to which the optical 
activity is due, may be represented by [C 6 H 7 05Cu]2[Cu(NH 3 )4], 
in which one atom of copper is equivalent to one CeHioOs residue. 2 



1 K. Hess and E. Messmer, Ber., 54 (1921), 834-841. 

2 K. Hess and E. Messmer, Ann., 435 (1923), 7. 



CHAPTER VII 
GELATINIZED CELLULOSE 

Gelatinized cellulose, as the name implies, refers to cellulose 
having a swollen, gelatinous appearance. It may be opaque, 
translucent, or transparent. When the fibrous structure is 
wholly or largely destroyed, it forms a dense, horny product on 
being dried en masse. Ultimate analysis shows it to be identical 
with normal cellulose. In contrast with the latter, it shows 
increased hygroscopicity, solubility in cellulose solvents, ease 
of hydrolysis, and adsorption of dyes and cold Fehling's solution. 
In general also, it is more readily esterified. The greater reactiv- 
ity of gelatinized cellulose may be referred in most cases to a 
greater surface exposure and not to a chemical change in the 
cellulose. 

The older term "hydrated" cellulose was based on the errone- 
ous belief that mercerization was accompanied by the addition 
of a molecule of water. Mercerized is not suitably descriptive, 
since in the strictest sense this refers to the technical operation 
of treating a cellulose fiber with caustic soda and stretching to 
produce a luster. A similar effect can be obtained with acids 
and salts. Schwalbe 1 has introduced the name "swollen" 
cellulose. This is not sufficiently definitive, since a cotton hair 
will increase 40 per cent in cross-sectional area by merely wetting 
with water. No name so far proposed is, however, entirely 
free from criticism. 

Gelatinized cellulose may be produced through the following 
agencies: (1) mechanical comminution and pressure in the 
presence of water; (2) salts; (3) bases; (4) acids; and (5) regenera- 
tion from esters. 

Gelatinized cellulose differs from hydrocellulose in having a 
lower copper number and a higher percentage of hygroscopic 

1 C. G. Schwalbe, "Ullmann's Enzyklopadie Tech. Chemie," 3 (1916), 
333. 

196 



GELATINIZED CELLULOSE 



197 



moisture. 1 The figures for various gelatinized celluloses are 
given below. 





Hygro- 


Copper 


Copper 




Substance 


scopic 
moisture, 


number 
before 


number 
after 


Difference 




per cent 


hydrolysis 


hydrolysis 




Cotton wool 


6.1 


1.1 


3.3 


2.2 


Cotton wool mercerized 




with: 










8 per cent NaOH 


7.7 


0.9 


3.2 


2.3 


16 per cent NaOH 


10.7 


1.3 


5.0 


3.7 


24 per cent NaOH 


11.3 


1.2 


6.1 


4.9 


40 per cent NaOH 


12.1 


1.9 


6.6 


4.7 


Glanzstoff silk 


9.8 
10.7 
10.2 
11.0 
11.4 

3.6 


1.5 
1.9 
3.0 

2.9 
4.1 

5.7 


12.8 
14.0 
14.5 
16.6 
17.7 
6.6 


11.3 


Viscose silk A 


12.1 


Viscose silk B 


11.5 


Viscose silk C 


13.7 


Chardonnet silk 


13.6 


Girard's hydrocellulose 


♦ 0.9 


Mitscherlich wood pulp, 










unbleached 




2.4 

2.8 


4.4 
3.5 


0.9 


Ritter-Kellner wood pulp . . 


2.7 



Coloration with Iodine. — Iodine reagents stain gelatinized 
cellulose, the color and its stability being dependent on the 
degree of gelatinization. Barreswil 2 observed that cellulose 
swollen by caustic soda and potash is colored blue by iodine. 
Lange 3 found that when ordinary cotton was stained with zinc 
chloriodine, the color was rapidly removed by washing with 
water, while under the same conditions mercerized cotton 
remained blue for a considerable time. It has been stated 4 
that gelatinized cellulose generally reacts with iodine in aqueous 
solution to give a blue color, though Htibner 5 was unable to 
obtain this color with mercerized cotton. 



1 C. G. Schwalbe, Z. angew. Chem., 22 (1909); 197; cf. Ibid., 20 (1907), 
2166. 

2 C. Barreswil, J. pharm. chim., 21 (1852), 205. 

3 H. Lange, Fdrber-Ztg., 14 (1903), 365. 

4 C. F. Cross and E. J. Bevan, "Cellulose," (1895), p. 7. 

5 J. Hubner, J. Soc. Chem. Ind., 27 (1908) 105. 



198 CHEMISTRY OF CELLULOSE AND WOOD 

A qualitative test of the degree of gelatinization may be 
obtained by staining the cellulose blue with zinc chloriodine 
and observing the rapidity with which the color is removed by 
washing with water. 1 Iodine-potassium iodide solution may be 
used, but it is less satisfactory. Celluloses having a high degree 
of gelatinization, such as mercerized cotton, parchment paper, 
and viscose silk, retain the blue color for a considerable time. 
The color disappears from Pauly silk (cuprammonium process) 
very rapidly. While cellulose gelatinized with strong chemical 
reagents may retain the blue color under washing for an hour, 
mechanically gelatinized cellulose loses its color almost 
immediately. 2 

The best solution for distinguishing between ordinary cotton 
and cotton mercerized with caustic soda is prepared as follows: 3 

A. Two-hundred and eighty grams of zinc chloride made up to 
300 cubic centimeters with water. 

B. One gram of iodine and 20 grams of potassium iodide in 100 
cubic centimeters of water. 

C. To 20 cubic centimeters of A are added 4 drops of B. The 
resulting solution C stains mercerized cotton a very dark, reddish- 
navy-blue shade, while ordinary cotton is faintly tinted reddish. 
Cotton that has been treated with saliva or mercerized with 
sulphuric, nitric, and syrupy phosphoric acids, saturated solu- 
tions of zinc chloride and sodium sulphide, and barium mercuric 
iodide also give the characteristic blue color. Cotton treated 
with hydrochloric acid and saturated solutions of potassium 
mercuric iodide, or potassium iodide, behaves like ordinary 
cotton. 

Hygroscopic and Colloid Water. — Cross and Be van 4 considered 
as hydrated those forms of cellulose containing water of consti- 
tution, but not of hydrolysis, as in the case of hydrocellulose. 
Hygroscopic moisture, the amount of which is dependent on 
atmospheric conditions, is removed at 100°, while water of 
hydration passes off at appreciably higher temperatures. To 

1 C. G. Schwalbe, Ber., 40 (1907) 4525: Z. angew. Chem., 20 (1907) 2169. 

2 C. G. Schwalbe, Z. angew. Chem., 32 (1919) 356. 

3 J. Hubner, J. Soc. Chem. Ind., 27 (1908), 107. 

4 C. F. Cross and E. J. Bevan, "Researches" (1901), p. 27; "Cellulose" 
(1905), pp. 23, 29. 



GELATINIZED CELLULOSE 



199 



designate the various forms of water, "aq" was proposed for 
that driven off at 100°, "Aq" for that evolved on further heating 
up to 150°, and H 2 for water actually retained through hydrol- 
ysis. 1 The evolution of water from gelatinized cellulose at 
various temperatures has been compared to the behavior of 
inorganic substances containing water of crystallization. 2 

Higgins 3 found that the hygroscopicity of mercerized cotton 
increased with the concentration of the sodium hydroxide 
employed. 





NaOH, 

specific gravity 


Moisture, 
per cent 


Ordinary cotton 




1.05 

1.10 

1.15 

1.20 

1.25 

1.30 

1.35 


6.2 


Mercerized with 


6.37 


Mercerized with 


6.68 


Mercerized with 


8.40 


Mercerized with 


9.41 


Mercerized with 


9.43 


Mercerized with 


9.57 


Mercerized with 


9.69 



Viscose in an atmosphere saturated with moisture at normal 
temperature will take up about 30 per cent of its original weight 
of water. 4 It is stated by Beadle 5 that cotton or viscose when 
finely ground is less hygroscopic than large pieces. Normal 
cotton contains 7 per cent of moisture in comparison with 4 per 
cent for the ground material. This is explained on the theory that 
hygroscopicity depends on internal stresses in the cellulose and 
that comminution reduces them. It has been shown recently 
that cellulose undergoes decided changes even with careful dry 
grinding. 6 The individual particles swell instantly and assume 
the shape of spheres on contact with water. The cellulose 



1 C. F. Cross and E. J. Bevan, Chem. Ztg., 33 (1909), 368. 

2 C. G. Schwalbe, Z. angew. Chem., 20 (1907), 2167; C. F. Cross, Ber., 
44 (1911), 153. 

3 S. H. Higgins, J. Soc. Chem. Ind., 28 (1909), 188. 

4 H. S. Mork, J. Franklin Inst, 184 (1917), 353. 

5 C. Beadle, J. Franklin Inst., 143 (1897), 12. 

6 H. Wislicenus and W. Gierisch, Kolloid Z., 34 (1924), 169. 



200 CHEMISTRY OF CELLULOSE AND WOOD 

becomes partially soluble in water, and shows a greater reduction 
of Fehling's solution and ease of hydrolysis as grinding progresses. 
The hygroscopic water, according to Schwalbe, 1 could be 
determined by heating at 100 to 105°. Distillation with kerosene 
gave the total water from which the hygroscopic water was sub- 
tracted to obtain the water of hydration. 

Water of 

Hydration, 

Per Cent 

Cotton cloth . 23 

Cotton, beaten to a pulp 1 . 56 

Cotton, mercerized and beaten to a pulp 6.31 

This distinction is considered invalid by Ost and Westhoff, 2 
since small progressive losses 3 of water are obtained by raising 
the temperature from 100 to 130°. In no case are fixed points 
obtained. Gelatinized cellulose contains more hygroscopic 
moisture than normal cellulose or hydrocellulose, and this can be 
determined accurately by gradually raising the temperature of 
the material, preferably in an atmosphere of dry hydrogen or 
carbon dioxide, to 125°. The hygroscopic moisture of mercer- 
ized cotton and hydrocellulose, determined at 130°, was 10.05 
and 4.73 per cent, respectively. Distillation with xylol gave no 
more water than drying at 120 to 125°. Mercerized cotton and 
viscose dried at the latter temperatures had exactly the same 
ultimate composition as normal cellulose. These conclusions 
have been vigorously disputed. 4 

Miller 5 had shown that a cellulose hydrate did not exist. 
Ordinary cellulose contained 7.12 per cent and mercerized 
cellulose 10.0 per cent of water at a temperature of 21.5 to 22.5°. 
When the mercerized cellulose was exposed to anhydrous cal- 
cium chloride, the water evaporated at a rate corresponding to 
an adsorbed and not a chemically combined substance. 

No satisfactory evidence has yet been presented to show that 
mercerized cellulose differs in chemical constitution or composi- 

1 C. G. Schwalbe, Z. angew. Chem., 20 (1907), 2166. 

2 H. Ost and F. Westhoff, Chem. Ztg., 33 (1909), 197; Ann., 382 (1911), 
354. 

3 c/. C. F. Cross and E. J. Bevan, "Cellulose" (1895), p. 4. 

4 C. F. Cross and E. J. Bevan, " Researches" (1905-1910), pp. 36-38. 

5 0. Miller, Ber., 43 (1910), 3430; 44 (1911), 728; cf. C. G. Schwalbe, 
Ibid., 44 (1911), 151; C. F. Cross, Ibid., 44 (1911), 153. 



GELATINIZED CELLULOSE 



201 



tion from normal cellulose, though various physical measure- 
ments show a difference. Ost 1 found that the viscosities of 
cuprammonium solutions of mercerized and untreated cotton 
were practically identical and concluded that no chemical change 
accompanied mercerization. Nakano, 2 however, found that 
cuprammonium solutions of mercerized cotton showed a decided 
decrease in viscosity. The cotton, 10 grams, was mercerized by 
treatment with 200 cubic centimeters of sodium hydroxide for 
24 hours. The relative viscosities of the cuprammonium solu- 
tions of cotton mercerized with various strengths of sodium 
hydroxide were: 



NaOH, per cent 


Viscosity 


NaOH, per cent 


Viscosity 



11 


4.22 
3.59 


17 
20 


3.03 
2.01 



It would seem that the act of dissolving in cuprammonium solu- 
tion might mask mercerization or produce effects similar to it. 

Gelatinized cellulose in many ways behaves as a hydrophilic 
colloid. It should, accordingly, be possible to obtain definite 
information as to the presence of " bound" water by applying 
the method of Newton and Gortner 3 for determining the hydro- 
philic colloid content of expressed plant fluids. The freezing- 
point depression, A, of the liquid is first obtained. The water 
content is then determined by any convenient method. To a 
weighed portion of the liquid, sufficient sucrose is added to make 
an exactly molar solution in the total water that is present, and a 
new freezing-point depression, A a, is then taken. In case no 
bound water occurs, the second freezing-point depression will be 
2.085° lower than A. When hydrophilic colloids are present 
there should be a depression greater than 2.085°, since some of the 
water, being bound by the colloids, is not free to dissolve the 
sugar. 

Theories of Gelatinization. — Gelatinization is here used in a 
narrower sense than hydration. The subject of hydration in the 



1 H. Ost, Z. angew. Chem., 24 (1911), 1892. 

2 M. Nakano, C. A., 16 (1922), 2405. 

3 R. Newton and R. A. Gortner, Botan. Gaz., 74 (1922), 442-446. 



202 CHEMISTRY OF CELLULOSE AND WOOD 

cacti has been extensively investigated by MacDougal. 1 He 

states : 

The source of energy in growth and swelling is the unsatisfied attrac- 
tion of molecules, or particles, or ions bearing an electrical charge. 
Substances made up in this manner may unite with definite proportions 
of water which become part of a symmetrical chemical structure, the 
union being known in classical chemistry as hydration. In addition, 
however, it is known that such particles may also adsorb and hold in 
combination additional molecules of water, an action especially charac- 
teristic of swelling in colloids, and the term hydration is used in the 
present work to include the entire range of action. 

According to Ostwald, 2 peat may hold water in five forms: 
(1) water of occlusion, held in spaces 1 millimeter and more in 
diameter; (2) capillary water; (3) colloidally bound water; (4) 
osmotically bound water; (5) chemically combined water, as in 
hydrocellulose, or water of crystallization, as in copper sulphate. 
The term " hydration" should be, accordingly, confined to the 
broad phenomena of the taking up of water in all its forms. 

It has long been realized that the gel state of cellulose is of 
great importance in the plant kingdom. The cellulose in young 
wood in comparison with old is much less resistant to chemical 
reagents. £ The amount of cellulose isolatable from green fodder 
plants is increased by a preliminary dehydration with alcohol. 4 

The phenomenon of " hydration" has been described 5 as a 
condition of intermolecular distention whereby the surface 
reactions are largely increased and the adsorption property cor- 
respondingly developed. 

The cause of gelatinization is not at all clear. Some of the 
theories that have been advanced to account for the phenomenon 
of the swelling of colloids, that have a bearing on the swelling of 
cellulose are: 

1. The capillary theory, in which it is assumed that swelling 
takes place by the passage of the liquid through small pores, the 

1 D. T. MacDougal, "Hydration and Growth," Carnegie Inst. Pub., 297, 
Washington (1920). 

2 W. Ostwald, Kolloid Z., 29 (1921), 318. 

3 C. G. Schwalbe, Z. angew. Chem., 22 (1909), 197. 

4 C. F. Cross and E. J. Bevan, "Cellulose" (1895), p. 7. 

5 J. F. Briggs, Ann. Rep. Soc. Chem. Ind., 2 (1917), 128. 



GELATINIZED CELLULOSE 203 

impelling force being the difference in surface tension between 
solvent and solution. 

2. The osmotic theory, 1 in which liquids pass into the sub- 
stance, the swelling being due to the difference in osmotic 
pressure between the external and internal solutions. Bartell 
and Sims 2 assume that in the swelling of a gel the latter consists 
of a framework of cells and that some unit of this framework 
functions as a permeable membrane. Swelling is due to imbi- 
bition — a process which is not well understood and which may 
be due to capillary action or to intermolecular reactions — 
together with the operation of effects that produce anomalous 
osmosis. The curves for swelling and osmosis show a close 
parallel. When a gel swollen in pure water is placed in a solution 
of an electrolyte an abnormal swelling effect would be due to 
negative osmotic action, and shrinking to an abnormally positive 
osmotic action. 

3. In the solid-solution theory of Katz, 3 colloidal substances 
capable of swelling are highly viscous liquids. The addition 
of a miscible liquid to such substances forms a solid solution. 

4. The electrostatic repulsion theory. 4 A gel consists of a 
sponge-like structure with many minute pockets filled with 
water. On the addition of acid or alkali, H or OH ions, respec- 
tively, are adsorbed on the surface of these pockets. The colloid 
particles become charged with electricity of the same sign. 
Owing to electrostatic repulsion, increase in size by imbibition 
of solution takes place. The addition of a neutral salt produces 
a decrease in swelling, due to neutralization of the original 
electrostatic repulsion. 

The cellulose fiber, according to Bovard, 5 consists of a sponge- 
like structure of colloid particles, strongly charged negatively, 
and held together by capillary attraction. This structure has 

1 W. Pfeffer, "Osmotische Untersuchungen," Leipzig (1877), 236 pp. 
C. P. Smith, J. Am. Chem. Soc., 43 (1921), 1350. 

2 F. E. Bartell and L. B. Sims, J. Am. Chem. Soc, 44 (1922), 289: 
F. E. Bartell, "Colloid Symposium Monograph," Madison, Wis. (1923), 
120. 

3 J. R. Katz, Z. physiol Chem., 96 (1916), 255; Kolloidchem. Beihefte, 
9 (1917), 182. 

4 R. C. Tolman and A. E. Stearn, J. Am. Chem. Soc., 40 (1918), 264. 

5 W. M. Bovard, Paper, 22, 3 (1918), 11-16. 



204 CHEMISTRY OF CELLULOSE AND WOOD 

the power of adsorbing hydroxyl ions and, since the latter are 
more abundant in an alkaline medium, gelatinization takes 
place most rapidly in a basic solution. The positive sodium 
or hydrogen ions, e.g., are drawn towards the cellulose particles 
by reason of the electrostatic attraction of the positive and nega- 
tive charges and, since they have like charges, they tend to repel 
one another. It is this pulling force or tendency to expand 
that causes the cellulose to swell with the simultaneous absorp- 
tion of water. A similar theory has been developed by Gesell 
and Minor. 1 

The adsorption of hydroxyl ions by cellulose, which has a 
marked residual valence, is greatly aided by mechanical treat- 
ment, such as beating in the presence of water. 2 The hydroxyl 
ions hydrolyze the cellulose molecule with increasing velocity 
into a series of products, the first being insoluble and mucilagi- 
nous, the later ones soluble dextrins or acids. The formation 
of cellulose "mucilage" is a hydrolytic reaction and must be 
distinguished from gelatinization, even though the two reactions 
may occur simultaneously. Knoevenagel 3 attributes the forma- 
tion of cellulose mucilage to the setting free of residual affinities 
by disintegration of the large cellulose molecule. Both acids 
and alkalis may set free residual affinities, thus making the 
modified cellulose more reactive. 

5. The diffusion-pressure theory of Proctor and Wilson. 4 
The force producing the swelling is due to the diffusion of the 
necessary excess concentration of diffusible ions of the gel phase 
over that of the external solution. Gelatin behaves as an ampho- 
teric colloid and forms highly ionized salts with acids and bases. 
The non-colloid ion on diffusing out exerts a pull that swells 
the gel. 

Of the ionized gelatin chloride the chloride ions remain in the solution 
in the interstices while their corresponding gelatin cations form part of 
the network and are not in solution in the same sense as the chloride 
ions. Because these chloride ions are balanced only by the positive 

1 W. H. Gesell and J. E. Minor, Paper, 24, 12 (1919), 15-19. 

2 J. E. Minor, J. Ind. Eng. Chem., 13 (1921), 131. 

3 E. Knoevenagel and H. Busch, Cellulosechemie, 3 (1922), 43. 

4 H. R. Procter and J. A. Wilson, J. Chem. Sac, 109 (1916), 307; J. A. 
and W. H. Wilson, J. Am. Chem. Soc, 40 (1918), 264. 



GELATINIZED CELLULOSE 205 

electrical charges on the gelatin network, there results an unequal dis- 
tribution of all ions between the external solution and that absorbed by 
the gelatin such that the total concentration of ions is always greater in 
the jelly than in the external solution. In tending to diffuse into the 
external solution the anions of the protein salt exert a pull upon the 
cations forming part of the gelatin network, causing an increase in 
volume of the jelly directly proportional to the excess of concentration 
of diffusible ions of the jelly over that of the external solution. 1 

Donnan 2 in his theory of membrane equilibrium was the first 
to stress the importance of the unequal distribution of diffusible 
crystalloidal ions on the swelling of colloids. His theory has 
been of the highest importance in explaining the chemistry of 
tanning, and it has been applied to the swelling of cellulose. 3 
It is improbable that any one theory can be made to fit all 
cases. The swelling of rubber in benzene might be explained on 
the solid-solution theory, while the other theories would fall down. 

It appears that ions are hydrated and that they carry water 
molecules with them. Hence, if the ions are adsorbed on 
the surface of the colloid particle, there is a concentration of 
water about the particle. 4 

According to Fischer, 5 hydration is a change by which the 
colloid enters into physicochemical combination with water. 
Solution, on the other hand, is the expression of the degree of 
dispersion of the colloid. The two are antagonistic, for finely 
dispersed colloid particles are incapable of binding as much water 
as coarse. 

The dilution at which some substances will form a gel is 
amazing. The glucoside apiin from parsley is stated not to 
lose its ability to gelatinize until 8500 parts of water have been 
added. 6 It is generally accepted that in gel formation the 
molecules unite to form interlacing filaments. In this way 
water or other liquid may be held by capillarity in addition to 

1 J. A. Wilson, "Colloid Symposium Monograph," Madison, Wis. (1923), 
p. 226. 

2 F. G. Donnan, Z. Elektrochem., 17 (1911), 572. 

3 J. E. Minor, Paper Trade J., 76, 23 (1923), 55-59. 

4 T. Svedberg, " Colloid Chemistry" (1924), p. 178. 

5 M. H. Fischer, Kolloid Z., 17 (1915), 2. 

6 E. von Gerichten, Ber., 9 (1876), 1122. 



206 CHEMISTRY OF CELLULOSE AND WOOD 

adsorption. A hot aqueous solution of caffeine on cooling forms 
a false gel of interlacing crystalline needles. Dibenzoylcystine 1 
forms a solid transparent gel containing 99.8 per cent of water. 
It has a fibrillar structure made up of exceedingly minute crystals. 

A solution of viscose on exposure to the air forms a network of 
interlacing cellulose filaments. 2 This may explain why a cellulose 
coagulum from viscose, containing 85 per cent of water, when 
dehydrated under high pressure, instead of forming a solid sheet, 
sometimes shows a laminated structure. 3 

Dyeing Properties. — Higgins 4 found a direct relation between 
the hygroscopicity of cotton and the adsorption of dyestufTs. 
Mercerized cotton does not dye to as full a shade after drying as 
before, the wet and dried fibers taking up 1.74 and 1.16 per cent, 
respectively, of benzopurpurin. 5 The change is apparently 
permanent, as it is not restored by exposure to air or prolonged 
steeping in water. Unmercerized cotton does not show this 
phenomenon. Cotton mercerized with hydrochloric acid of sp. 
gr. 1.185 to 1.19 likewise took up more dye if washed and dyed 
previous to drying. 6 

The affinity of cellulose for dyestuffs is proportional to the 
colloidal activity produced by the moisture, i.e., colloid water, 
normally present. 7 This affinity is reduced by dry heat and 
steam, but can be maintained or increased during such treatments 
by the presence of hygroscopic substances such as glycerine. 

The increased affinity of mercerized cotton for dyes is due to 
the higher colloid state of the cellulose. 8 It requires from 40 to 50 
per cent less of substantive dyestuffs than ordinary cotton to 
obtain the same depth of color. 9 

Cotton treated with sodium hydroxide of a sp. gr. of only 1.005 
showed an increased affinity for the direct dye benzopurpurin 

1 R. A. Gortner and W. F. Hoffman, J. Am. Chem. Soc, 43 (1921), 2199. 

2 H. Rousset, Rev. gen. sci. pur. appl, 20 (1909), 831. 
3C. Beadle, Chem. News, 71 (1895), 213. 

4 S. H. Higgins, J. Soc. Chem. Ind., 28 (1909), 188. 

5 E. Knecht, J. Soc. Dyers Colourists, 24 (1908), 107-109. 

6 E. Knecht, Chem. Trade J., 56 (1915), 45. 

7 E. Justin-Mueller, Fdrber-Ztg., 24 (1913), 98. 

8 W. P. Dreaper, Chem. News, 90 (1904), 179. 

9 W. Schaposchnikoff and W. Mjnajeff, Z. Farben und Texiilchem., 
3 (1904), 163; 4 (1905), 81. 



GELATINIZED CELLULOSE 207 

4B. 1 With solutions up to a sp. gr. of 1.09 the increase in affinity 
is roughly proportional to the concentration of the alkali; from 
sp. gr. 1.09 to 1.11 the increase in affinity is greater than the 
corresponding increase in the concentration of the alkali, this 
effect reaching a maximum at sp. gr. 1.13 to 1.15. 

Gelatinized cellulose is dyed slightly or not at all by basic 
dyes. 2 

Gelatinization with Salts. — The use of zinc chloride for 
producing parchment paper was patented 3 in England in 1859. 
Paper was treated with a 76 per cent solution of the salt at a 
temperature of 27 to 100°. Its action is weaker than that of 
76 per cent sulphuric acid. 4 

The use of zinc chloride for mercerizing was mentioned by 
Mercer in his original patent. Grandmougin 5 states that a 
solution of 72° Be. produces, at 40 to 50°, effects similar to mercer- 
ization with sodium hydroxide; it is claimed by Minajeff, 6 how- 
ever, that cotton treated with a zinc chloride solution of 64° Be. 
(60.5 per cent) does not show a noticeably increased affinity for* 
dyes. Cotton immersed in a 50 per cent solution of zinc chloride 
showed a shrinkage of 2.5 per cent. 7 Stretching produced a 
slight luster. 

In the manufacture of vulcanized fiber, advantage is taken of 
the gelatinizing action of zinc chloride. Paper made from 
cotton is passed through a bath of zinc chloride of 70° Be. at a 
temperature of 40°. It is then rolled up on heated drums to 
obtain the desired thickness, the heat causing gelatinization. 
The material is then leached to a chlorine content of less than 
0.15 per cent. This process requires 3 to 4 weeks for material 
0.25 inch thick, and 6 to 8 months for that 2 inches thick. The 
wet fiber is dried at 40 to 60°, pressed, and calendered. The 
shrinkage during drying and finishing is 50 per cent. The fin- 
ished product has a sp. gr. of 1.1 to 1.48. Vulcanized fiber has a 

1 J. Hubner and W. J. Pope, J. Soc. Chem. Ind., 23 (1904), 404-411. 

2 C. G. Schwalbe, Z. angew. Chem., 20 (1907), 2172. 

3 Thomas Taylor, Dinglers polytech. J., 155 (1860), 397. 

4 J. Ferwer, Dinglers polytech. J., 159 (1861), 221. 

5 E. Grandmougin, Bull. Mulhouse, 68 (1898), 348; Chem. Ztg., 32 (1908), 
241. 

6 W. Minajeff, Z. Farben-Ind., 9 (1910), 65. 

7 J. Hubner and W. J. Pope, J. Soc. Chem. Ind., 23 (1904), 409. 



208 CHEMISTRY OF CELLULOSE AND WOOD 

low resistance to water. In 24 hours it will take up from 25 to 
70 per cent. 1 Red vulcanized fiber on prolonged contact with 
water absorbed 0.76 cubic centimeter in 34 days, and 0.62 cubic 
centimeter in 19 days, per cubic centimeter of fiber. 2 The 
increase in volume was 161 and 151 per cent respectively. 
Water at a temperature of 60° was absorbed more rapidly than 
cold, but the amount absorbed and the increase in volume was 
less than at ordinary temperature. 

Cellulose swells and becomes transparent in a saturated 
solution of sodium mercuric iodide. It forms a horny mass 
after being washed and dried. 3 

The mercerizing action of iodine salts was discovered by 
Hubner. 4 A concentrated solution of potassium iodide has 
about the same mercerizing action as a concentrated solution of 
sodium hydroxide. Washing with alcohol removed the 
potassium iodide completely. Saturated solutions of barium 
iodide and potassium mercuric iodide also produce mercerization. 

According to von Weimarn, 5 any salt having a moderately 
high solubility in water can peptize 6 cellulose, provided certain 
conditions, such as concentration, pressure, temperature, and 
duration of action, are observed. In carrying out the peptiza- 
tion, 3 grams of cellulose, such as filter paper, and 100 cubic 
centimeters of water are placed in a vessel and heated, the salt 
being gradually added. Agitation hastens the conversion of the 
cellulose to a gelatinous plastic condition. With certain salts, 
such as calcium, barium, and strontium sulphocyanates, sodium, 
strontium, and calcium iodides, and calcium bromide, the 
process can be carried out by heating at atmospheric pressure. 
Sodium, potassium, and barium chlorides require heating at 
higher pressures. Peptization with a saturated solution of 

1 C. Almy, Met. Chem. Eng., 13 (1915), 746. 

2 F. H. Parker, Phil. Mag., 25 (1913), 210. 

3 A. G. Duboin, Compt. rend., 141 (1905), 385-388. 

4 J. Hubner and W. J. Pope, J. Soc. Chem. Ind., 22 (1903), 70-77; 23 
(1904), 409. 

5 P. P. von Weimarn, Z. Chem. Ind. Kolloide, 11 (1912), 41-43; Kolloid- 
Z., 29 (1921), 197-198, 198-199; G. P. 275882 (1912). 

6 Graham defined peptization as the conversion of a gel into a sol by means 
of a small quantity of a dispersing agent. In this chapter peptization is used 
in the broader sense of increasing the distance between particles. 



GELATINIZED CELLULOSE 209 

sodium chloride begins at a pressure of about 8 atmospheres, 
corresponding with a temperature of 170°. 

Schwalbe 1 found that pure cellulose heated with a solution of 
sodium chloride at 180° for 3 hours did not swell or show indica- 
tion of solution. He concludes that von Weimarn must have 
used an impure cellulose, such as absorbent cotton or filter paper. 

Peptization of cellulose proceeds rapidly with boiling saturated 
solutions of LiCl, LiBr, Lil, LiN0 3 , Nal, Srl 2 , Sr(CNS) 2 , CaBr 2 , 
Cal 2 , Ca(CNS) 2 , Ba(CNS) 2 , and Mn(CNS) 2 . With salts that 
produce a very rapid peptization, such as the thiocyanates, a 
1 per cent solution of cellulose forms a transparent jelly on 
cooling. Prolonged heating leads to degradation of the cellulose. 
Swelling and dispersion, at ordinary temperatures and pressures, 
proceed in the following order: NaCl, BaCl 2 , KI, SrCl 2 , CaBr 2 , 
and LiCl. There is no noticeable swelling with sodium chloride 
after 5 years. 

According to Herzog and Beck, 2 the dispersion of cellulose 
in strong solutions of the salts of the caustic alkalis and alkaline 
earths is a function of the hydration of the ions of the particular 
salt. Their effectiveness increases in the following order: NH 4 , 
K, Na, Li, Ba, Sr, Ca, ^S0 4 , CI, Br, I, CNS. Neither swelling 
nor solution takes place in the chlorides, bromides, nitrates, 
sulphates, thiocyanates, acetates, and lactates of Na, K, NH 4 , 
and Mg; the bromides of Ba and Sr; or barium thiocyanate. 
Calcium thiocyanate is a very efficient solvent, though this 
property is not possessed by all thiocyanates. 

A 30 per cent solution of sodium sulphide produced a shrinkage 
of 1.3 per cent in cotton. Stretching gave a slight luster. 3 

Adsorption of Salts. — Schonbein 4 observed that when strips of 
filter paper were dipped into solutions of acids, bases, and salts, 
the water always reached a greater height than the chemicals. 
When filter paper is spotted with aqueous solutions of the metallic 
salts, the area of concentration is greatly influenced by the kind 
of salt and paper, the temperature, and the concentration of the 

1 C. G. Schwalbe, Farber-Ztg., 24 (1913), 433. 

2 R. O. Herzog and F. Beck, Z. physiol. Chem., Ill (1920), 287-292; cf. 
F. Beck, Z. angew. Chem., 34 (1921), 113-114. 

3 J. Hubner and W. J. Pope, J. Soc. Chem. Ind., 23 (1904), 409. 

4 C. F. Schonbein, Chem. Zentr., 32 (1861), 881-884. 



210 CHEMISTRY OF CELLULOSE AND WOOD 

solution. 1 With strong solutions of copper sulphate the salt 
spreads all over the spot, while with dilute it concentrates in the 
middle. Cadmium salts have a marked tendency to spread over 
the entire wetted area. 

In slightly acid solution the metal ion of the salts of the heavy 
metals rises as high as the water, though this is not the case 
with neutral solutions. 2 The adsorptive power of filter paper is 
due mainly to the fact that is has the property of a coarse, disperse 
medium as the result of its distensible capillaries. 3 Cellulose 
behaves as a negative colloid, flocculating and adsorbing the 
positive colloids such as the sols of metallic hydroxides. The 
adsorption of neutral salts is very slight. 

When vegetable fibers are treated with alkaline plumbate 
or plumbite solutions, lead oxide is fixed by them. The 
hydroxides of Cu, Zn, Co, and Fe'" can also be fixed directly. 
Cotton treated with Schweizer's reagent retains only traces of 
cupric oxide after washing; if the reagent contains sodium or 
potassium hydroxide, however, cupric oxide is fixed by the fiber. 4 
In dilute cuprammonium solution, cotton shows a high preferen- 
tial sorption of copper. 5 This may be due to chemical action. 

When 1 gram of cellulose was immersed in 50 cubic centimeters 
of 1 per cent ferric chloride solution, filter paper fixed 0.229 per 
cent of iron, and cotton 0.112 per cent. Lower results were 
obtained with 0.1 per cent solution. With 1 per cent solutions 
of mercuric chloride and mercuric acetate, cotton fixed 0.2 
and 1.5 per cent of mercury, respectively. 6 Cotton is capa- 
ble of adsorbing and fixing small amounts of metal from 
dilute solutions of copper, nickel, aluminum, and potassium 
salts. Unbleached cotton shows this property to a greater 
extent than bleached. 7 The fixation is reduced to a minimum 
by treating the cotton with dilute sodium hydroxide and washing 

1 T. Bayley, J. Chem. Soc, 33 (1878), 304-306; cf. Mansier, J. pharm. 
chim,, [6] 16 (1902), 60-64, 116-120. 

2 1. M. Kolthoff, Pharm. Weekblad, 57 (1920), 1515. 

3 K. Scherinja, Pharm. Weekblad, 57 (1920), 1289. 

4 A. Bonnet, Compt. rend., 117 (1893), 518-519; 121 (1895), 700-701. 

5 H. F. Coward, L. Spencer, and F. C. Wood, J. Textile Inst., 14 (1923), 
529. 

6 W. Schellens, Arch. Pharm., 243 (1905), 617-631. 

7 B. Rassow, Z. angew. Chem., 24 (1911), 1127. 



GELATINIZED CELLULOSE 211 

thoroughly. The phenomenon is shown to a greater extent by 
the artificial silks, and not at all by hydrocellulose. The adsorp- 
tion of the salt solutions is independent of time, temperature, and 
concentration. 

The source and purity of the cotton or other form of cellulose 
is of great influence. Haller 1 found negative adsorption with 
aluminum acetate, and in some cases with aluminum sulphate. 
With aluminum acetate, negative adsorption was least with raw 
cotton and increased with the purification of the fiber. With 
aluminum sulphate, the adsorption was positive with cotton in 
the raw state and after boiling with lime; however, boiling with 
caustic soda and bleaching caused negative adsorption. The 
adsorption of lead acetate was positive in all cases and increased 
with the purity of the fiber. Adsorption and fixation of the 
insoluble base do not necessarily run parallel. In every case 
there was a fixation of aluminum oxide with aluminum salts, 
even though negative adsorption values were sometimes obtained. 
According to Heermann, 2 the adsorption values recorded by 
Haller should be considered as "apparent adsorptions" consist- 
ing of the summation of adsorption and secondary reactions. 
The latter are sufficiently important in some cases as to show an 
apparent negative adsorption. 

Durst 3 found that the fixation of alumina was influenced by 
such factors as the previous treatment of the cotton, time of 
immersion, concentration of the solution, temperature, and 
ratio of solution to cotton. Using bleached cotton and aluminum 
acetate, the adsorption of alumina was 0.064 per cent, as deter- 
mined from the ash content of the cotton, and only 0.047 per 
cent calculated on the change in concentration of the solution. 

When sulphite pulp was treated in a metal beater with 3 per 
cent of its weight of aluminum sulphate, the salt was quantita- 
tively decomposed, all the alumina being fixed by the fiber. 4 
Sutermeister 5 worked with poplar soda pulp, spruce sulphite 
pulp, and rag stock in glass. He found that the adsorption of 

1 R. Haller, Chem. Ztg., 42 (1918), 597; 43 (1919), 195. 

2 P. Heermann, Chem. Ztg., 43 (1919), 195. 

3 G. Durst, Chem. Ztg., 43 (1919), 374. 

4 C. G. Schwalbe and H. Robsahm, Wochbl. Papierfabr., 43 (1912), 1454. 

5 E. Sutermeister, Pulp Paper Mag. Can., 11 (1914), 803. 



212 CHEMISTRY OF CELLULOSE AND WOOD 

alumina was independent of the concentration of the aluminum 
sulphate solution and that the amount adsorbed was in no case 
greater than 0.3 per cent of the weight of the fiber. 

Tingle 1 has studied the same problem, using as cellulose 
acid-washed filter paper and bleached spruce sulphite pulp. 
With both neutral and basic solutions of aluminum sulphate, 
no change in the aluminum content of the solutions could be 
detected, except when a pulp contained calcium compounds and 
the wash water gave a distinctly alkaline reaction. 

These findings have been recently confirmed by Schwalbe 2 
in a reinvestigation in which non-metallic vessels were used. 
Neither cotton nor wood pulp adsorbed appreciable amounts of 
alumina from solutions of aluminum sulphate. 

Cellulose in the colloidal condition produced by prolonged 
beating is capable of fixing metallic bases from aluminum sul- 
phate and other salts. 3 Beaten hydrocellulose and oxy cellulose 
possess this property to a high degree. The fixation of dyestuffs 
and metallic bases of. neutral salts during the steaming or aging 
of mordanted materials is probably due to the formation of 
hydrocellulose and oxycellulose. The difficulty in wetting 
materials mordanted with metallic salts is caused by an irreversi- 
ble state of the colloidal cellulose in combination with the metallic 
hydroxide. 

Bancroft 4 states that all aluminum salts hydrolyze more or 
less readily in aqueous solution. The hydrolysis is greater with 
the weaker acids and increases with rising temperature. Cotton 
adsorbs alumina effectively only in basic solutions. Only 
colloidal alumina is taken up and firmly held by the fiber, coagu- 
lated alumina being held mechanically. It is probable that in 
all cases alumina and not a basic salt is adsorbed, though the 
phenomenon is complicated by the fact that alumina itself will 
adsorb some sulphuric acid. The increased adsorbing power of 
mercerized cotton is probably due to a structural difference in the 
fiber. 

1 A. Tingle, J. Ind. Eng. Chem., 14 (1922), 198. 

2 C. G. Schwalbe, Z. angew. Chem., 37 (1924), 125. 
3 C. G. Schwalbe, Z. angew. Chem., 32 (1919), 355. 

4 W. D. Bancroft, J. Phys. Chem., 26 (1922), 501-536. 



GELATINIZED CELLULOSE 213 

Kolthoff 1 found that practically no adsorption of metals took 
place from solutions of the alkali and alkaline earth salts. True 
adsorption of electrolytes by filter paper does not take place. 2 
The phenomenon is a purely chemical one and depends on the ash 
content of the filter paper. The ash functions as a permutite: 
Ca-permutite + M + + <=± M-permutite + Ca ++ . The metal may 
be Pb, Cu, Hg, Cd, Mn, or Fe. Silver, cobalt, and nickel are 
adsorbed to a less extent, and the alkali and alkaline earth ions 
not at all. This explains the prevalent belief that cellulose 
behaves as an acid and combines with metallic oxides. 

Filter paper, according to Kolthoff, 3 showed negative 
adsorption of chlorine ions from sodium chloride. Masters 4 
found that when pure neutral cellulose is extracted with sodium 
chloride and other neutral salts, an acid extract is obtained, while 
extraction of the cotton with water gives an alkaline solution. 
Extraction of 10 grams of cotton with a 20 per cent solution of 
sodium chloride gave 3.9 milligrams of hydrogen chloride. After 
washing the cotton free from alkali and chlorine ions, it again 
gave an acid extract with salt solutions. Temperature had no 
effect on the alkalinity and acidity obtainable. The occurrence 
of hydrochloric acid in the filtrate obtained by passing a sodium 
chloride solution through purified cotton had been previously 
observed by Harrison. 5 

Schwalbe 6 records a case where a gelatinized cotton pulp 
contained colloidal copper derived from the bronze knives in the 
beater. The pulp dyed bluish black, in place of the normal red 
color, with benzopurpurin 10B. 

Gelatinization with Alkalis. Mercerization. — In a broad way, 
mercerization applies to the treatment of cellulose fibers with 
chemical reagents that produce swelling followed by shrinkage 
to dimensions below the original, on removal of the reagents. 
Technically, mercerization refers to the treatment of the fibers 
with strong caustic soda, the fibers being held under tension 

1 1. M. Kolthoff, Pharm. Weekblad, 58 (1920), 94. 

2 I. M. Kolthoff, Pharm. Weekblad, 57 (1920), 1515; 58 (1921), 152, 233. 

3 I. M. Kolthoff, Pharm. Weekblad, 58 (1921), 94. 
4 H. Masters, J. Chem. Soc, 121 (1922), 2026. 

5 W. Harrison, J. Soc. Dyers Colourists, 27 (1911), 286. 
« C. G. Schwalbe, Z. Chem. Ind. Kolloide, 2 (1908), 229. 



214 CHEMISTRY OF CELLULOSE AND WOOD 

during the alkaline treatment or while the alkali is being removed 
by washing. Fibers so treated show a greatly increased luster 
when dry. 

The process is named from John Mercer, who was the first 
at least to appreciate its industrial possibilities. Mercer showed 
great aptitude for experimentation and was to the English 
dyeing trade of his day what Caro later became for the German 
dyestuff industry. In 1850 he took out a patent, 1 in which is 
stated : 

My invention consists in subjecting vegetable fabrics to the action of 
caustic soda or caustic potash, dilute sulphuric acid, or chloride of zinc, 
of a strength and temperature sufficient to produce new effects. I pass 
the cloth through a padding machine charged with caustic soda or caus- 
tic potash at 60 or 70° Twaddel's hydrometer, at the common 
temperature. 

Mercer observed that the fabrics after removal of the alkali 
showed decided shrinkage, increased strength, firmness, and 
affinity for dyes. The ability of strong sodium hydroxide to 
produce shrinkage merely had been previously known in France. 2 
Mercer, being apparently satisfied with the increased solidity 
and strength of his fabrics, did not employ tension ; hence, luster 
was not apparent. In all justice it must be said that Thomas 
and Prevost, 3 in 1896, were the first to appreciate the silky luster 
attainable by mercerizing under tension, though the essentials of 
the process had been specified by Lowe. 4 He, however, was most 
concerned with preventing loss by shrinkage. Luster is not 
obtained unless the fibers are under tension during the alkaline 
treatment or while the alkali is being washed out. 

When immersed in strong sodium hydroxide, the normally 
flat, twisted cotton fiber undergoes a great change. It swells 
to a cylindrical form, untwists, and contracts considerably in 
length. At the expense of the length, the walls become thicker 
and the lumen nearly disappears. The swelling in sodium 

1 John Mercer, E. P. 13296 (1850). 

2 J. Persoz, "Traits Theorique et Pratique de 1' Impression des Tissus," 
IH (1846), pp. 139-141. 

3 R. Thomas and E. Prevost, E. P. 20714 (1896). 

4 H. A. Lowe, E. P. 4452 (1890); cf. P. and C. Depoully, E. P. 28696 
(1883); H. A. Lowe, E. P. 20314 (1889). 



GELATINIZED CELLULOSE 215 

hydroxide of a concentration of 15 to 20 per cent is sufficient to 
rupture the cuticle. The resulting bulges give a dumb-bell 
appearance. Technically, solutions containing 27 to 32 per cent 
of sodium hydroxide have been found most satisfactory for 
mercerization. In water, cotton swells in cross-section 40 to 42 
per cent, and this is increased to 280 per cent in 20 per cent 
sodium hydroxide. 1 

Paper made from cotton mercerized with 23 per cent sodium 
hydroxide, in comparison with untreated cotton, showed a 
decrease in stretch and tensile strength. 2 While the strength 
of the individual fibers is increased by mercerization, this treat- 
ment converts them into smooth cylinders that slip past one 
another, under tension, more easily than the normal twisted 
cotton fiber. Mercerization greatly increases the absorbency 
of the paper due to removal of the water-repellent incrustants. 

Cotton increases in tensile strength from 30 to 50 per cent 
through mercerization. The increase is less when merceriza- 
tion takes place under tension. 3 Krais 4 studied the shrinkage 
of cotton when treated with sodium hydroxide solutions of various 
strengths at 0, 10, and 20°. The maximum shrinkage of 31.3 
per cent occurred with a solution of sp. gr. 1.320, and at a tem- 
perature of 10°. 

Up to a concentration of SN, about the limit of solubility, 
lithium hydroxide produces greater shrinkage than equivalent 
concentrations of sodium and potassium hydroxides. The 
latter solutions produce about the same shrinkage up to 5N; 
between 5N and 8iV, sodium hydroxide is more active than potas- 
sium hydroxide, but at greater concentrations potassium 
hydroxide is the more active. Tetramethyl-ammonium 
hydroxide produces considerable shrinkage at high concentra- 
tions, but none is obtained with ammonium hydroxide, hydra- 
zine hydrate, calcium hydroxide, and barium hydroxide. 5 

1 R. S. Willows and A. C. Alexander, J. Textile Inst, 13 (1922), 237. 

2 C. G. Schwalbe, K. Fenchel, and R. Cornely, Wochbl. Papierfabr., 44 
(1913), 757-759. 

3 R. S. Willows, T. Barratt, and F. H. Parker, J. Textile Inst, 13 
(1922), 229. 

4 P. Krais, Z. angew. Chem., 25 (1912), 2649. 

6 E. Knecht and W. Harrison, J. Soc. Dyers Colourists, 28 (1912), 224. 



216 CHEMISTRY OF CELLULOSE AND WOOD 

While sodium hydroxide, due to its greater viscosity, has less 
mercerizing action than potassium hydroxide at high concentra- 
tions, under ordinary conditions 1 gram-molecule of sodium 
hydroxide is equivalent in its action to 1.5 gram-molecules of 
potassium hydroxide. 1 

The heat liberated during mercerization increases with the 
concentration of the sodium hydroxide but is not proportional 
to it. 2 There are two breaks in the curve, the first between 10 
and 15 per cent sodium hydroxide and the second at about 30 
per cent sodium hydroxide. 

Mercerization appears to be complete in about 3 minutes. 3 
It is stated to be instantaneous and that immersion for a period 
longer than 1 minute is useless. 4 Between 40 and 180 seconds, 
there is not much difference in mercerization in so far as the 
dyeing capacity of the cotton is concerned. 5 

While various theories have been advanced to account for the 
luster of mercerized cellulose, it is generally accepted that this 
is due to the production of a smooth surface from which the 
light is reflected, 6 and to translucency of the cell wall. The more 
nearly round the fiber, the greater is the luster. 7 Hubner and 
Pope 8 found that the untwisting of the ribbon-like cotton fiber 
is of great importance. They attributed the luster to a series 
of spiral ridges that reflect the light. This is improbable, since 
fibers mercerized with just enough tension to keep them straight 
are without luster. 

The alkali-cellulose fiber, on washing with water, becomes 
inelastic and cannot be stretched to its original length without 
breaking. If washed with ammonia, however, the fiber retains 
its elasticity. This also becomes inelastic on washing with 
water. If the fiber is under tension while the ammonia is being 



1 E. Ristenpart, C. A., 15 (1921), 2728. 

2 T. Barratt and J. W. Lewis, J. Textile Inst., 13 (1922), 113. 

3 R. S. Willows et at, I.e. 

4 F. J. G. Beltzer and J. Persoz, "Les Matieres Cellulosiques " (1911), p. 
62. 

6 E. Knecht, J. Soc Dyers Colourists, 24 (1908), 68. 

6 H. Lange, Fdrber-Ztg., 9 (1898), 197; W. Harrison, /. Soc. Dyers 
Colourists, 31 (1915), 198; T. Barratt, J. Soc. Chem. Ind., 43 (1924), 593B. 

7 A. Adderly, /. Textile Inst., 16 (1924), 195T. 

8 J. Hubner and W. J. Pope, J. Soc. Chem. Ind., 23 (1904), 410. 



GELATINIZED CELLULOSE 217 

washed out, it acquires a luster. 1 This is probably explainable 
on the observation of Coward and Spencer 2 that there is marked 
hysteresis in the shrinkage of cotton fiber on replacing solutions 
of mercerizing strength with weaker ones. 

Vieweg 3 found that cellulose absorbs more alkali from a solution 
of sodium hydroxide saturated with sodium chloride than with- 
out it. Whether this is true or not appears to depend upon the 
temperature and the particular concentration of the solution in 
sodium hydroxide and salt. 4 According to Miller, 5 sodium 
chloride and sodium carbonate in small amounts are without 
effect. It is well established, however, that the presence of 
sodium chloride reduces the shrinkage and luster, 6 and the 
affinity for dyestuffs; 7 hence the degree of gelatinization. 
Hubner 6 observed the following differences: 

Shrinkage, 
Per Cent 

Sodium hydroxide, 12 per cent 11.5 

Sodium hydroxide, 12 per cent, saturated with NaCl. . . 6.8 

Sodium hydroxide, 20 per cent 18.6 

Sodium hydroxide, 20 per cent, saturated with NaCl. . . 15.7 

Sodium silicate 8 also reduces the mercerizing action of sodium 
hydroxide. 

Alkali-cellulose. Adsorption of Alkalis. — There has been 
much dispute as to whether sodium hydroxide forms a definite 
compound with cellulose. Many investigators have failed to 
find evidence of a chemical compound. 9 Through its reaction 

1 E. Thiele, Chem. Zlg., 25 (1901), 610. 

2 H. F. Coward and L. Spencer, /. Textile Inst., 14 (1923), 32. 

3 W. Vieweg, Ber., 41 (1908), 3269. 

4 C. Beadle and H. P. Stevens, Eighth Int. Cong. Appl. Chem., 13 (1912), 
37; J. Soc. Dyers Colourists, 30 (1914), 244. 

5 O. Miller, J. Russ. Phys.-Chem. Soc, 37 (1905), 361; Ber., 41 (1908), 
4302. 

6 J. Hubner, J. Soc. Chem. Ind., 28 (1909), 228; W. Herbig, Fdrber-Ztg., 
23 (1912), 136. 

7 E. Knecht, J. Soc. Chem. Ind., 28 (1909), 228. 

s J. Hubner and W. J. Pope, J. Soc. Chem. Ind., 23 (1904), 409; W. 
Herbig, Fdrber-Ztg., 23 (1912), 136; P. Gardner, " Mercerisation der 
Baumwolle" (1898), p. 44. 

9 C. Beadle and H. P. Stevens, J. Soc. Dyers Colourists, 30 (1914), 244; 
A. Leighton, J. Phys. Chem., 20 (1916), 32; H. F. Coward and L. Spencer, 
J. Textile Inst., 14 (1923), 32. 



218 



CHEMISTRY OF CELLULOSE AND WOOD 



with chromic acid mixture, Thies 1 deduces that no alkali-cellulose 
compound is formed. Compounds such as sodium ethylate 
give decidedly more carbon dioxide on oxidation in a given time 
than a simple mixture of alkali and alcohol. Gladstone's soda- 
cellulose by this test does not behave differently from pure 
cellulose. 

Karrer 2 maintains that a definite addition compound, Ci 2 H 2 oOio.- 
NaOH, remains after treating cellulose with concentrated 
sodium hydroxide and washing with alcohol. The more recent 
papers are in entire agreement with this view. 3 According to 
Vieweg, 4 alkali-cellulose behaves precisely as an alcoholate in 
its reactions with carbon bisulphide and benzoylchloride. Gela- 
tinization evidently produces some change in the cellulose 
molecule. Vignon 5 found the following heats of combustion for 
precipitated and mercerized cellulose: 

Calohies 

Purified cotton (dried at 110°) 4,223 

Precipitated cellulose (from Schweizer's reagent) 3 , 982 

Mercerized cellulose 3 , 980 

Lowering of the heat of combustion was believed to be due to 
depoly merization . 

Mercerization causes "depolymerization" of the cellulose 
molecule with decomposition into water and carbon dioxide on 
long standing. 6 Mercerized sulphite cellulose prepared for 
viscose showed the following changes : 



Cellulose, 
per cent 



NaOH, 

per cent 



Na 2 C0 3 , 
per cent 



H 2 0, 

per cent 



Alkali-cellulose, fresh 

Alkali-cellulose, after standing 
3 years at 15° in the absence 
of air 



24.48 



19.40 



15.36 



6.40 



1.72 



7.95 



58.44 



66.25 



1 F. H. Thies, Fdrber-Ztg., 24 (1913), 393. 

2 P. Karrer, Cellulosechemie, 2 (1921), 125. 

3 W. Vieweg, Z. angew. Chem., 37 (1924), 1008-1010; E. Heuser, Ibid., 
1010-1013; F. Dehnert and W. Konig, Cellulosechemie, 5 (1924), 107-112; 
E. Knecht and J. H. Platt, J. Soc. Dyers Colourists, 41 (1925), 53-55. 

4 W. Vieweg, Zellsioff u. Papier, 2 (1922), 18. 

5 L. Vignon, Compt. rend., 131 (1900), 708. 

6 H. Eggert, Kunststoffe, 3 (1913), 381. 



GELATINIZED CELLULOSE 219 

The cellulose had decomposed to the extent of losing its fibrous 
structure. 

Cellulose mercerized with 17 per cent sodium hydroxide under 
tension, when examined by the X-ray method, gives a point 
diagram like the original fiber. If the fiber is not under tension, 
Debye-Scherrer rings, not a point diagram, result, showing 
that the crystallites occur in every possible position in the fiber. 
Any slight difference between the original fiber and that mer- 
cerized under tension may be due to the migration of an oxygen 
bridge in the cellulose molecule. 1 Gebhard 2 has suggested that 
mercerization results in enolization. 

Gladstone 3 treated cotton with strong sodium hydroxide and 
washed it with cold absolute alcohol or hot alcohol of sp. gr. 
0.825. The alkali remaining in the fiber agreed with the formula 
Ci 2 H 2 oOi .NaOH and was considered as combined. Thiele, 4 
by the arbitrary method of a limited washing with strong 
ammonia of cotton treated with 30 per cent sodium hydroxide, 
concluded that the compound Ci 2 H 2 oOi .2NaOH was formed. 
Cross and Bevan 5 state that a definite reaction takes place in 
the molecular ratio Ci 2 H 20 Oi :2NaOH. The latter becomes 
Ci 2 H 20 Oi .NaOH on washing with alcohol. 

Vieweg, 6 in 1907, shook cotton with solutions of sodium 
hydroxide of various strengths and determined the amount of 
alkali absorbed by the cotton, by titrating a portion of the 
liquid. The curve obtained on plotting the amount of sodium 
hydroxide absorbed against the concentration of the original 
solution showed two definite breaks. The first break was 
obtained at concentrations of 16 to 24 grams of sodium hydrox- 
ide per 100 cubic centimeters of solution, corresponding with an 
absorption of 13 per cent of alkali by the cellulose and a molecular 
ratio of (C 6 Hio0 5 )2: NaOH. The second break was obtained at 
concentrations of 35 grams and greater per 100 cubic centimeters, 

1 R. O. Herzog and G. Londberg, Ber., 57 (1924), 329. 

2 K. Gebhard, Chem. Ztg., 37 (1913), 663. 

3 J. Gladstone, J. Chem. Soc., 5 (1853), 17. 

4 E. Thiele, Chem. Ztg., 25 (1901), 610. 

5 C. F. Cross and E. J. Bevan, " Cellulose" (1895), p. 23. 

6 W. Vieweg, Ber., 40 (1907), 3876-3883; cf. Ibid., 41 (1908), 3269-3275. 



220 CHEMISTRY OF CELLULOSE AND WOOD 

corresponding with an absorption of 24 per cent of sodium 
hydroxide and a molecular ratio of (C 6 Hi O 5 )2: 2NaOH. 

Miller 1 worked in essentially the same way as Vieweg but 
instead obtained a smooth curve. He claimed that the Glad- 
stone compound did not exist and that the action of strong 
sodium hydroxide on cotton was a solid-solution phenomenon. 2 
Htibner and Teltscher 3 likewise came to the conclusion that 
"soda-cellulose" as a definite compound does not exist. Hot 
absolute alcohol was found to remove practically all of the sodium 
hydroxide. The results of Gladstone were due to his failure to 
wash the cotton with cold absolute alcohol until equilibrium 
had been reached. Their own results on adsorption, wherein 
the cotton was washed with alcohol, have been criticized on 
the ground that the amount of caustic soda removed by the 
alcohol is a function of the temperature; 'and further that the 
method gives no definite indication of the true adsorption in 
aqueous solution. 

The question was carefully investigated by Leigh ton, 4 who 
obtained smooth adsorption curves and found no evidence 
whatever of a definite chemical compound. If cotton is shaken 
with alkali, a portion of the supernatant liquid analyzed, and 
the amount of adsorption calculated from the decrease in con- 
centration, the adsorption figure is not accurate unless no 
liquid is taken up by the cotton. Obviously, this is not possible. 
A direct or gravimetric method of determining the adsorption 
will give too high results unless all the unadsorbed solution is 
removed. If the cotton takes up water and alkali in the same 
ratio as they exist in the solution, no adsorption occurs. 
Leighton shook cotton with alkali for 3 hours and to avoid 
obvious errors centrifuged it for 1 hour at 4000 r.p.m. 

Below a concentration of 90 grams of sodium hydroxide per liter, 
no adsorption takes place. For a concentration of 300 grams per 
liter there is obtained an adsorption of 0.27 gram of sodium 
hydroxide per gram of cotton by analyzing the solution, and an 

1 O. Miller, Ber., 40 (1907), 4903; 41 (1908), 4297; Chem. Ztg., 29 (1905), 
491. 

2 O. Miller, Ber., 43 (1910), 3430. 

3 J. Hubner and F. Teltscher, J. Soc. Chem. Ind., 28 (1909), 641-644. 

4 A. Leighton, /. Phys. Chem., 20 (1916), 32-50. 



GELATINIZED CELLULOSE 
Adsorption of NaOH by Cotton 



221 



Volume of solution = 100 cubic centimeters; weight of cotton = 1.0 to 
1.2 grams; time of run = 3 hours 



Solution, 
grams 
NaOH 

per liter 


Grams NaOH 

adsorbed per gram 

cotton 


Solution, 
grams 
NaOH 

per liter 


Grams NaOH 

adsorbed per gram 

cotton 


Gravi- 
metric 


Titra- 
metric 


Gravi- 
metric 


Titra- 
metric 


474 
450 
427 
415 
400 
379 
350 
323 
300 
276 
275 
231 


0.916 

0.887 
0.875 

0.846 

0.782 

0.733 

0.665 


0.28 

0.27 
0.27 
0.27 
0.25 


230 
212 
210 
190 
189 
184 
175 
140 
138 
125 
90 
40 


0.664 
0.642 

0.594 
0.579 
0.572 

0.461 
0.433 


0.19 

0.12 
0.08 

0.01 
0.00 



adsorption of 0.75 gram by analyzing the centrifuged cotton. 
There is thus a marked difference between the titration curve 
and the gravimetric curve. The latter is much more accurate 
than the former, though the true adsorption curve must lie 
between them. 

According to Kolthoff, 1 cellulose takes up sodium hydroxide 
and potassium hydroxide from solutions up to 2N, in accordance 
with the partition law. Between 2N and QN the amount taken 
up increases greatly but is independent of the concentration; 
while between 6iV and 87V there is a further decided increase ; it is 
likewise independent of the concentration. He concludes that 
adsorption in the true sense does not take place but that chemical 
combination occurs. 

Harrison 2 attributes failure to follow the ordinary adsorption 
formula to the change in the physical state of the fiber produced 

1 1. M. Kolthoff, Pharm. Weekblad, 58 (1921), 46-56. 
2 W. Harrison, J. Soc. Dyers Colourists, 31 (1915), 198. 



222 CHEMISTRY OF CELLULOSE AND WOOD 

by certain concentrations of caustic soda. In the case of regen- 
erated cellulose, maximum absorption of water and caustic 
soda takes place in a 9 per cent solution of sodium hydroxide, 
while with cotton a 17.5 per cent solution is required. This is due 
to the resistance to swelling of the cotton cuticle, a condition 
obtaining until the concentration of the alkali is sufficient to 
cause it to burst. 1 Mercerization does not destroy the cuticle. 2 

The effect of mechanical hindrance in studying adsorption is of 
great importance. While there are certain objections to its use, 
regenerated cellulose should be the most suitable material. 
Cotton loose and in the form of fabric, absorbs the same amount of 
water; however, loose cotton absorbs about twice as much 
sodium hydroxide as the fabric. 3 The swelling of cotton fibers in 
solutions of sodium hydroxide is about equal to the volume of 
liquor absorbed, the limit of swelling being set by the cuticle. 
In sodium hydroxide solutions up to 14 per cent, both sodium 
hydroxide and water are absorbed rapidly and according to the 
strength of the solutions. Above 14 per cent, the cotton con- 
tinues to take up sodium hydroxide, but less rapidly, while there is 
a distinct decrease in the absorption of water. 

The "hydration" and sodium hydroxide absorption were 
determined by Beadle and Stevens 4 with regenerated cellulose 
(cuprammonium). Skeins were immersed in sodium hydroxide 
and weighed after removal of the surface liquid. The weight 
of alkali, determined by titration, subtracted from the tot#l 
increase in weight of the skeins gave the "hydration" or water 
absorbed. The data given in the table below represent combined 
adsorption and absorption phenomena. They are interesting 
in that they show how much water and sodium hydroxide are 
taken up by cellulose under certain conditions. The maximum 
amounts of both substances taken up are attained by a 9.0 per 
cent solution of sodium hydroxide at a temperature of 5°. 

1 C. Beadle and H. P. Stevens, J. Soc. Dyers Colourists, 30 (1914), 244. 

2 W. Minajeff, Z. Farben-Ind., 7 (1908), 1, 21. 

3 H. F. Coward and L. Spencer, J. Textile Inst., 14 (1923), 32. 

4 C. Beadle and H. P. Stevens, Eighth Int. Cong. Appl. Chem., 13 (1912), 
25. 



GELATINIZED CELLULOSE 



223 



Per cent 
NaOH in 


Hydration, cellulose = 
100 


NaOH absorption, cellulose = 
100 


bath 


At 5° 


At 20° 


At 40° 


At 5° 


At 20° 


At 40° 


1 


217 


182 


167 


3 


3 


3 


2 


279 


217 


192 


8 


8 


8 


3 


324 


241 


216 


16 


14 


14 


4 


426 


280 


224 


24 


16 


16 


5 


615 


300 


230 


42 


22 


17 


6 


1,380 


310 


238 


83 


25 


20 


7 


1,960 


380 


240 


150 


36 


25 


8 


2,576 


562 


261 


224 


56 


30 


9 


2,699 


758 


338 


256 


74 


38 


10 


1,800 


920 


440 


210 


98 


56 


11 


1,483 


861 


458 


182 


110 


66 


12 


1,310 


719 


480 


170 


112 


72 


13 


1,200 


620 


412 


161 


102 


76 


14 


1,003 


558 


360 


142 


96 


78 


15 


798 


500 


334 


132 


88 


74 


16 


762 


458 


310 


128 


84 


72 


17 


715 


420 


300 


124 


82 


71 


18 


658 


400 


280 


122 


82 


70 


20 


590 


360 


240 


110 


84 


72 


22 


540 


325 


220 


114 


86 


73 


25 


461 


280 


220 


120 


86 


76 



It is an old observation 1 that when solutions of the hydroxides 

of barium, strontium, and calcium are passed through filter paper 

N 
adsorption of the base takes place. Cellulose in a -^ solution of 

barium hydroxide at 0° took up 3.82 per cent of the base; in a 

N 

tt: solution of strontium hydroxide at 0°, 2.18 per cent. 2 

Vignon 3 found that 100 grams of cotton fixed 2.67 grams of 
barium hydroxide from a 1.0 per cent solution, and 0.39 gram 
from a 0.1 per cent solution. 



1 A. MtiLLER, J. prakt. Chem., 83 (1861), 384; H. Weiske, Landw. Vers. 
Sta., 19 (1876), 155. 

2 H. Wichelhaus and W. Vieweg, Ber., 40 (1907), 441. 
3 L. Vignon, Compt. Rend., 143 (1906), 550. 



224 



CHEMISTRY OF CELLULOSE AND WOOD 



Kolthoff 1 obtained no adsorption with sodium and potassium 
carbonate solutions up to 0.5N. Ammonium hydroxide was 
taken up in small quantity from solutions up to6iV, but independ- 
ently of the concentration. Barium hydroxide was taken up 
in amounts proportional to the square roots of the final 
concentrations. 

Grimm, 2 in studying the action of caustic alkalis and alkaline 
earths, found that the greatest adsorption takes place in 4 to 6 
hours, after which there is a decrease. The greater the concentra- 
tion of the solution the greater is the adsorption. With some 
bases motion increases adsorption while with others it decreases 



Bleached cotton 



Per cent adsorbed based on weight of fiber 



CaO 



SrO 



BaO 



K 2 



Na 2 



Standing 
Shaking . 



2.04 

2.83 



4.08 
2.52 



0.72 
0.48 



0.63 
0.32 



it. The alkaline earths were strongly adsorbed, the caustic alkalis 
only weakly so. 

Ammonia is retained by cellulose fibers. 3 Filter paper 4 in 
an atmosphere containing ammonia will take up detectable 
amounts. Cotton will take up 115 times its volume of ammonia 
at atmospheric pressure. 5 Dry ammonia is adsorbed by dry 
cellulose 6 in approximately the ratio NH3:4C 6 Hi O 5 . 

Vignon 7 heated ammonium hydroxide of 22° Be. and cellulose 
in sealed tubes. Between 100 and 200° combination took place 
with the fixation of 1.05 to 2.86 per cent of nitrogen. The 
presence of an amino group in the cellulose was indicated by 
an affinity of the compound for acid dyes, and by the heat of 
reaction with sulphuric acid. Ammonia at elevated tempera- 
tures readily attacks the artificial silks, especially Chardonnet, 

1 1. M. Kolthoff, Pharm. Weekblad, 58 (1921), 46. 

2 H. Grimm, Zellstoff u. Papier, 1 (1921), 33-56. 

3 Mansier, J. pharm. chim., 16 (1902), 61. 

4 H. R. Procter, J. Soc. Chem. Ind., 23 (1904), 9; L. L. de Koninck, 
Bull. soc. chim. Belg., 23 (1909), 221. 

5 F. H. Bowman, "The Structure of the Cotton Fibre," London (1908), 
p. 201. 

6 B. Oddo, Gazz. chim. ital, 49, II (1909), 127. 

7 L. Vignon, Compt. rend., 112 (1891), 487. 



GELATINIZED CELLULOSE 



225 



reducing it to a powder having the properties of aminoeellulose. l 
The formation of an amino group requires confirmation. 

The action of dilute sodium hydroxide at various tempera- 
tures has been studied by Schwalbe and RobinofT. 2 This is 
very slight for pure cotton but becomes considerable with chemi- 
cally modified material. The cotton was purified by heating 
with alkaline rosin soap, washing with hot water, and carefully 
bleaching. It had a corrected copper number of 0.04. With 
concentrations up to 7 per cent, a 4 per cent solution of the 
alkali had the most drastic action at room temperature. The 
copper number of the cotton so treated was 0.257. The solu- 
bility decreased with increase in concentration, becoming inappre- 
ciable with 9.0 per cent sodium hydroxide. The solubilities were: 





NaOH solution, per cent 




1 


2 


3 


4 


5 


6 


7 


8 


9 


Solubility, per 
cent 


0.74 


0.53 


0.49 


0.42 


0.34 


0.30 


0. 


14 


0. 


06 




The copper numbers of the cotton treated 
hydroxide at various temperatures were: 


with 


sodium 



Temperature 



Strength of NaOH solution, per cent 



Copper numbers 



Room. 
100... 
135... 
150... 
179... 
213... 



0.150 
0.180 
0.142 
0.100 



0.166 
0.200 
0.170 
0.150 
0.110 



0.195 
0.262 
0.395 
0.280 
0.128 



0.257 
0.528 
0.890 
0.700 
0.445 



0.135 
0.168 
0.285 
0.120 
0.050 
0.000 



Cold and hot alkali up to 100° has little effect, but above this 
point the action increases with rise in temperature. Judging 
from the copper number, 4 per cent sodium hydroxide has the 

1 F. J. G. Beltzer and J. Persoz, "Matieres Cellulosiques," p. 104. 

2 C. G. Schwalbe and M. Robinoff, Z. angew. Chem., 24 (1911), 256-258. 



226 



CHEMISTRY OF CELLULOSE AND WOOD 



greatest effect at elevated temperatures also. The gum values, 
i.e., the amount of flocculent material precipitated by alcohol 
after neutralization of the alkaline extracts, increased greatly 
above 150°, so that this appears to be the critical temperature. 
The alkali presumably decomposes the cellulose with the forma- 
tion of substances related to hydrocellulose. 

Tests for Mercerization. — The use of iodine solutions has 
been mentioned. The reaction with substantive dyes is useful. 
Ordinary cotton dyed with benzopurpurin 4B is turned blue 
immediately by hydrochloric acid while mercerized cotton 
becomes reddish violet. 1 On reduction by heating with titanous 
chloride, ordinary cotton appears indigo blue and mercerized 
cotton, red just prior to complete destruction of the color. To 
obtain this difference the cotton must be treated with caustic 
soda of a specific gravity greater than 1.15; if stretched during 
mercerization, the specific gravity must be over 1.175. Knaggs 2 
found that reduction is not necessary. If strong hydrochloric 
acid is added to the dilute dye bath, in which the samples are 
immersed, until the shade of the unmercerized sample just 
becomes blue, the mercerized cotton remains a bright red. 

The degree of mercerization 3 can be determined by dyeing the 
cotton with benzopurpurin 4B, the amount of dyestuff taken up 
being estimated by indirect titration. 4 Hanks, treated under 
tension in caustic soda solutions of various strengths, were dyed 
together in the same bath. The fibers then contained the 
following amounts of dye: 



Specific gravity, 


Dye, 


Specific gravity, 


Dye, 


NaOH solution 


per cent 


NaOH solution 


per cent 


1.05 


1.88 


1.225 


3.27 


1.10 


2.39 


1.25 


3.38 


1.125 


2.57 


1.275 


3.50 


1.15 


2.95 


1.30 


3.56 


1.175 


3.02 


1.325 


3.60 


1.20 


3.15 


1.35 


3.66 



1 E. Knecht, J. Soc. Dyers Colourists, 24 (1908), 67. 

2 A. B. Knaggs, J. Soc. Dyers Colourists, 24 (1908), 112. 
3 E. Knecht, J. Soc. Dyers Colourists, 24 (1908), 68. 

4 E. Knecht, J. Soc. Dyers Colourists, 21 (1905), 3. 



GELATINIZED CELLULOSE 



227 



Ordinary cotton took up 1.77 per cent of the dye. Thus, the 
amount of dye taken up augments with the increase in the 
strength of the caustic soda. 

Miller 1 found that increased adsorption of dyes began with 
cotton treated with 9 per cent caustic soda. The adsorption 
increased to a maximum lying beyond 25 per cent caustic soda. 

Vieweg 2 has proposed a method for testing mercerization based 

on the adsorption of dilute alkali. One gram of the material 

is shaken in a wide-mouthed bottle with 100 cubic centimeters of 

N 

■n sodium hydroxide for 30 minutes, and a portion of the liquid 

N 



withdrawn and titrated with — acid. 



Ordinary cotton adsorbs 

1.0 per cent of sodium hydroxide. The adsorption reaches 2.8 
with cotton that has been mercerized with 16 per cent sodium 
hydroxide, but does not exceed 2.9 with cotton mercerized with 
50 per cent alkali. Some forms of cellulose show a higher adsorp- 
tion. Obviously, the " adsorption" is not only influenced by the 
degree of gelatinization but by the extent to which the alkali may 
react chemically with the modified celluloses. 



NaOH, 
per cent 



NaOH, 

per cent 



Pure cellulose 
Sulphite pulp . 

Soda pulp 

Sulphate pulp 



1.0 
1.2 

1.6 
1.7 



Filter paper 

Regenerated cellulose, 
cuprammonium 

Viscose 



1.6 
4.0 
4.5 



Briggs 3 has shown that the adsorption is much greater in 

alcoholic solutions of sodium hydroxide and increases with the 

strength of the alcohol. Artificial silks show an adsorption as 

N 
high as 14 to 15 per cent, based on the cellulose, in -^ sodium 



1 O. Miller, Ber., 43 (1910), 3430; cf. J. Hubner and W. J. Pope, J. Soc. 
Chem. Ind., 23 (1904), 404. 

2 W. Vieweg, Ber., 40 (1907), 3876; Chem. Ztg., 32 (1908), 329; Papier- 
Ztg., 34 (1909), 1352. 

3 J. F. Briggs, Chem. Ztg., 34 (1910), 455. 



228 CHEMISTRY OF CELLULOSE AND WOOD 

hydroxide solution in 93 per cent alcohol. With the latter 
solution, it is possible to show differences in the degree of 
gelatinization attained by beating pulp in a Hollander. 

Gelatinization with Acids. — The parchmentizing action of 
strong mineral acids has been discussed elsewhere (Chap. IX). 
While they have a mercerizing effect, the luster is not compar- 
able to that obtained with the caustic alkalis. The mercerized 
product is harder, and without due care gelatinization is soon 
followed by hydrolysis. 

Mercer 1 used sulphuric acid of 105° Tw. (sp. gr. 1.525 or 62 
per cent) at a temperature not exceeding 50°F. Thomas and 
Prevost 2 specify acid of sp. gr. 1.518 to 1.62. Sulphuric acid of 
sp. gr. 1.526 to 1.611, and phosphoric acid of sp. gr. 1.70 give to 
cotton, with the exception of luster, properties very similar to 
those obtained by mercerization with sodium hydroxide. 3 

In addition to sulphuric acid, Hubner and Pope 4 worked with 
phosphoric, nitric, and hydrochloric acids. Concentrated hydro- 
chloric acid and nitric acid of sp. gr. 1.415 produced a contraction 
of the cotton of 9.5 per cent. A slight luster was obtained by 
tension. Cold hydrochloric acid, sp. gr. 1.16 to 1.18, has prac- 
tically no effect on cotton fiber. 5 It is strongly attacked by acid 
of sp. gr. 1.195 to 1.20. Acid of sp. gr. 1.185 to 1.19 shrinks 
the fiber and greatly increases its affinity for dyes. The specific 
gravity of an acid reading 1.185 is reduced to 1.18, indicating 
that a preferential adsorption of hydrogen chloride takes place. 

Cotton treated with nitric acid of sp. gr. 1.41 shows many of 
the properties of mercerization. 6 Mercerizing action does not 
begin at the ordinary temperature with acid below a sp. gr. of 
1.38. The most suitable strength is 1.415. Acids of sp. gr. 1.11 
to 1.31 have a tendering action. 7 



1 J. Mercer, I.e. 

2 R. Thomas and E. Prevost, E. P. 18040 (1895). 

3 E. Grandmougin, Z. Farben-Ind., 6 (1907), 2; cf. Chem. Ztg., 32 (1908), 
241. 

4 J. Hubner and W. J. Pope, J. Soc. Chem. Ind., 23 (1904), 409. 

5 E. Knecht, J. Soc. Dyers Colourists, 31 (1915), 8; Chem. Trade J., 56 
(1915), 45. 

6 E. Knecht, /. Soc. Dyers Colourists, 12 (1896), 89. 

7 E. Knecht, Ber., 37 (1904), 549. 



GELATINIZED CELLULOSE 229 

Adsorption of Acids. — Girard 1 treated hemp fibers with a very 
dilute solution of hydrochloric acid and dried them at 45°. 
They were then heated at 140° in a glass tube through which 
passed a current of air. The latter did not render a solution of 
silver nitrate turbid. The fibers were, however, converted into 
brittle hydrocellulose, from which all the acid was subsequently 
recovered. Cellulose may be tendered by acids, and yet it is 
very difficult to detect them. 2 

Sulphuric acid can be readily removed from cellulose by washing, 
while the more volatile acids, such as hydrochloric, nitric, sul- 
phurous, and acetic, are not only resistant to washing but to 
drying as well. Koechlin 3 likens this behavior to the adsorption 
of gases in a porous substance such as charcoal. Washing with 
water will not remove even small quantities of acetic acid from 
cotton, so that it is neutral to litmus. 4 To accomplish this end 
the moist cotton must be treated for an hour with superheated 
steam at a temperature of about 200°. 

Acid is more rapidly removed from cotton by washing it with 
a solution of sodium chloride than with pure water. 5 

Vignon 6 found that when cotton was immersed in 1.0 and 0.1 
per cent sulphuric acids, all the acid could be completely removed 
by washing with water. 

It is well known that filter paper will take up acids existing 
in the atmosphere of the laboratory. Cellulose dried over fused 
calcium chloride and treated with dry hydrogen choride in 
anhydrous ether, after washing with anhydrous ether, showed 
an adsorption of 0.46 per cent of hydrogen chloride. 7 

Cotton exposed to a stream of nitric oxide for 24 hours took 
up 2.40 per cent of the gas. 8 It retained 1.67 per cent nitric 
oxide after exhaustion by exposure in a desiccator containing 
sulphuric acid and a dish with ferrous sulphate and lime. 

1 A. Girard, Ann. chim. phys., [5] 24 (1881), 356. 

2 J. Hubner, J. Soc. Chem. Ind., 34 (1915), 935. 

3 C. Koechlin, Bull. Mulhouse, 58 (1888), 547. 

4 W. Zanker and O. Mann, Fdrber-Ztg., 27 (1916), 355. 

5 H. F. Coward and G. M. Wigley, J. Textile Inst, 13 (1922), 121. 
6 L. Vignon, Compt. rend., 143 (1906), 550. 

7 H. Jentgen, Z. angew. Chem., 29 (1910), 1544. 

8 T. Panzer, Z. physiol. Chem., 93 (1915), 388. 



230 



CHEMISTRY OF CELLULOSE AND WOOD 



Oddo 1 found that dry hydrogen chloride combined with 
cellulose in the definite ratio of one molecule of the gas to every 
(C 6 Hi O 5 )4. The other gases tested did not show this relation- 
ship. They were taken up in the order of their solubility in 
water, namly, HC1, H 2 S, N 2 0, C0 2 , C 2 H 2 , CO, O, N, CH 4 , and 
H. The order of absorption is comparable to that with charcoal. 
Nitrated cellulose (13.32 per cent nitrogen) absorbed about five 
times the volume of sulphur dioxide taken up by cellulose. 
Since the hydrogen chloride could not be displaced by an indif- 
ferent gas, it is assumed that a compound is formed of the nature 
of an oxonium derivative (I), or one similar to that obtainable 



O 



II 
R.CH:0 + HC1 = R.CHCl.OH 



R 



CI 



with aldehydes (II). 

Kolthoff, 2 working with aqueous solutions of hydrochloric, 
sulphuric, acetic, and oxalic acids in concentrations of 0.1 N or 
less (0.5iV or less for oxalic acid), found that the amount of the 
acid taken up by filter paper was almost exactly equivalent to 
the alkalinity of the ash of the paper. 

Hydrochloric, phosphoric, and sulphuric acids in moderate 
concentrations do not form compounds with cellulose at the 





Adsorption of HC1, grams per 100 grams of 


Solution, grams HC1 per 


cellulose 


liter 










Gravimetic 


Titrametric 


45.8 


2.0 





114.5 


3.1 





209.2 


6.2 


2.34 


288.9 


6.9 


7.00 


331.3 


7.3 


9.28 


373.6 


9.1 


11.63 


414.0 




13.97 



1 B. Oddo, Gazz. chim. ital., 49, II (1919), 127. 

2 I. M. Kolthoff, Pharm. Weekblad, 57 (1920), 1571. 



GELATINIZED CELLULOSE 231 

ordinary temperature. Leighton 1 found that selective adsorp- 
tion was most marked with hydrochloric acid and less with 
sulphuric acid. There was none with phosphoric acid. The 
procedure followed in this work was similar to that used with 
sodium hydroxide. A solution of sulphuric acid up to a con- 
centration of 400 grams per liter is absorbed as such. According 
to Coward, 2 cotton shows a preferential sorbtion of acid from 
a dilute solution of sulphuric acid. 

Gelatinization by Beating. — Soon after sulphite pulp became 
an article of commerce, it was found that prolonged beating pro- 
duced a slimy mass suitable for making "pergamyn," i.e., 
imitation parchment and greaseproof papers. 3 If grinding or 
beating is continued until the fibrous structure is destroyed, a 
4 per cent suspension has the consistency of thick honey. Evap- 
oration of the water at a temperature below 40° leaves a hard, 
horny mass known as "cellulith." 4 Its density varies from 1.4 
to 1.5. 

Gelatinization in a beater is due to both a brushing and a 
cutting action. Mere comminution produces a more apparent 
than real gelatinization; hence it is advisable to use dull knives 
that have a pounding rather than a cutting action. The pound- 
ing and brushing effects are increased by having the stock in the 
beater of the highest practical concentration. This is about 4 
per cent for the ordinary Hollander. The rate of gelatinization 
is slow at first and then increases rapidly. 

Fibers that have been completely dried gelatinize or "hydrate" 
less readily than those that have been preserved in the wet state. 
This is particularly true of ground wood, as is shown by the 
difficulty frequently encountered in separating the fibers in the 
beater. Some "hydration" takes place on the prolonged stand- 
ing of pulp in water, but the extent can be but very slight. It 
is stated, however, that in the manufacture of newsprint where 
the machine operates at speeds of 600 to 1000 feet per min- 

1 A. Leighton, J. Phys. Chem., 20 (1916), 188-194. 

2 H. F. Coward, L. Spencer, and F. C. Wood, J. Textile Inst., 14 (1923), 
529T. 

3 S. Ferenczi, Z. angew. Chem., 12 (1899), 51. 

4 H. Brunswig, U. S. P. 622325 (1899); C. D. Abel, E. P. 18930 (1897); 
M. M. Rotten, G. P. 98201 (1897). 



232 CHEMISTRY OF CELLULOSE AND WOOD 

ute, the extra hydration resulting from allowing the pulp to stand 
in the stock tanks for 24 hours, is sufficient to cause sticking to 
the presses of the machine. l This change may be due to bacterial 
action. 

In Germany the gelatinized Mitscherlich sulphite pulp used 
in the manufacture of pergamyn is known as cellulose mucilage 
(Zellstoffschleim). Hofmann 2 attempted to find a chemical 
difference between pergamyn pulp and the wood pulp from which 
it was prepared. It was found that the removal of the "wood 
gum" with 5 per cent sodium hydroxide did not interfere with 
the preparation of pergamyn. The two pulps contained equal 
quantities of water-soluble constituents, the pergamyn slightly 
less pentosans and methylpentosans than the wood pulp. He 
concluded that wood gum plays no part in gelatinization. The 
pergamyn was as resistant to acid hydrolysis, as measured by the 
sugar produced in a definite time, as the original pulp; hence 
mucilaginous carbohydrates having a lowered resistance to 
hydrolysis are not formed. He, accordingly, concluded that 
"hydration" or hydrolysis did not take place in the beater and 
that gelatinization was due to subdivision of the fibers into 
minute fibrilla3. 

Schwalbe 3 found that gelatinized cellullose does show a 
hydrolysis difference, but to obtain definite results it is necessary 
that there be a wide difference in the degree of gelatinization. 

Chemical wood pulp lends itself far more readily to mechanical 
disintegration than cotton, but has only a slight tendency to form 
fibrils. The coniferous fiber consists of an inner layer of spirally 
arranged strands and an outer layer that does not split. After 
prolonged beating the microscope shows the outer layer as 
irregular fragments, and the inner layer as a tangled mass of 
cloudy, gelatinous filaments. 4 

The gelatinization of cellulose by beating, according to Briggs, 5 
is not a purely mechanical phenomenon but a physicochemical one. 
A cellulose hydrogel is formed through molecular surface disten- 

1 J. A. de Cew, J. Soc. Chem. Ind., 36 (1917), 357. 

2 H. Hofmann, Z. angew. Chem., 20 (1907), 746-749. 

3 C. G. Schwalbe and E. Becker, Z. angew. Chem., 32 (1919), 268. 

4 E. Kirchner, Wochbl. Papier jabr., 35 (1904), 3411-3414. 

6 J. F. Briggs, Papier-Fabr. Fest u. Auslandsheft, 8 (1910), 46-49. 



GELATINIZED CELLULOSE 233 

tion, water being adsorbed and becoming what is now termed 
colloidal water. That water plays an essential part is shown by 
the failure of cellulose to gelatinize on beating in alcohol or 
petroleum. Furthermore, cellulose gelatinized by beating shows 
a marked reduction in the viscosity of its solutions, the 
viscosity decreasing with the degree of gelatinization. 

It appears that the gelatinization of cellulose by prolonged 
beating is accompanied and assisted by the formation of hydro- 
cellulose and oxy cellulose. 1 This is shown by the increase in 
reducing power. It is, in fact, possible to isolate from highly 
beaten pulps dextrin-like bodies having a strong reducing action 
on Fehling's solution. 

According to Schwalbe, gelatinization increases the adsorption 
of copper from cold Fehling's solution, to give a high cellulose 
number 2 . The increase in the cellulose number is probably due 
more to the swelling of the incrustants than to the cellulose 
itself. 3 The determination of the "hydrate" copper number is 
satisfactory for pure cellulose but not for determining the degree 
of gelatinization of ground wood or raw wood pulps, since the 
incrustants have a reducing action on Fehling's solution. 4 
According to Siebert and Minor, 5 the cellulose number of beaten 
wood pulps is no larger than the experimental error in the 
determination. 

The gelatinizing capacity of wood pulps is perceptibly 
decreased by long boiling with water or by steaming, 6 but is 
increased by a preliminary treatment with acids 7 or with oxidizing 
agents. 8 It has been shown that the high copper number of 
hydrocellulose is due to the presence of dextrin-like products that 
can be removed by extraction with water. 9 The time required 

1 C. G. Schwalbe, Z. angew. Chem., 32 (1919), 355. 
2 C. G. Schwalbe, "Chemie der Cellulose" (1911), 634. 

3 C. G. Schwalbe and E. Becker, Z. angew. Chem., 32 (1919), 265. 

4 C. G. Schwalbe, Papier-Fabr., 21 (1923), 77. 

5 F. B. Siebert and J. E. Minor, Paper 24, 23 (1919), 15. 

6 C. G. Schwalbe and E. Becker, Z. angew. Chem., 32 (1919), 265-269. 

7 C. G. Schwalbe, G. P., 303498 (1916); 323745 (1918); Rufferwahnt, 
G. P. 312179 (1918). 

8 C. G Schwalbe, G. P. 303305 (1917); 319826 (1918). 

9 O. Hauser and H. Herzfeld, Chem. Ztg., 39 (1915), 689; W. Harrison, 
J. Soc. Dyers Colourists, 28 (1912), 224. 



234 



CHEMISTRY OF CELLULOSE AND WOOD 



to bring a sulphite pulp of a low copper number to standard 
slowness by beating was reduced from 3.75 hours to 20 minutes 
by a preliminary trituration of the pulp in a mortar with 1 per 
cent hydrochloric acid. 1 Excessive acid treatment destroys the 
property of gelatinizing. 

The use of oxidizing agents appears to have been anticipated 
by Lowe. 2 Though probably unaware of the nature of the effect 
produced, he recognized fully the essentials for gelatinization. 
The fibers were first treated with a bleach solution and then 
beaten with dull knives so as to have a large surface, thus pound- 
ing, instead of cutting, the fiber, "to make the stuff hold 
the water." 

The effect of various chemicals in accelerating gelatinization, 
as determined by the increased bursting strength of the paper 
made therefrom, has been examined by Mansfield and 
Stephenson. 3 They used bleached sulphite pulp and a ball mill. 





Bursting 

strength, 

pounds 




Bursting 

strength, 

pounds 


Blank 

1 per cent Na2C0 3 

3 per cent Na 2 C0 3 

5 per cent Na 2 C03 

10 per cent Na 2 C0 3 

5 per cent NaOH 


51.5 

77.4 
86.0 
68.6 
88.4 

72.2 


5 per cent NH 3 

10 per cent NH 3 

5 per cent ZnCl 2 

10 per cent NaCl 

5 per cent H2SO4 

15percentCH 3 .COOH. . 


73.6 
84.2 
70.4 
52.0 
51.4 
50.6 



Zinc chloride and the bases in all cases caused a marked increase 
in gelatinization. The acids had no effect. It is recommended 
that during the beating operation 1 to 5 per cent of sodium 
carbonate, based on the pulp, be used. 

The beneficial effect of sodium carbonate and sodium hydroxide 
has been confirmed by Sutermeister. 4 He treated sarrples of 



1 F. B. Siebert and J. E. Minor, Pulp Paper Mag. Can., 18 (1920), 939. 

2 H. Lowe, U. S. P. 33092 (1861). 

3 E. K. Mansfield and J. N. Stephenson, Pulp Paper Mag. Can., 13 
(1916), 325. 

4 E. Sutermeister, Paper 23, 14 (1918), 11-13. 



GELATINIZED CELLULOSE 235 

steam-dried, bleached aspen soda pulp, and air-dried, bleached 
sulphite pulp with water and various chemicals for a period of 1 
to 3 weeks. The soluble materials were then removed by washing 
with water and the wet pulp ground in a ball mill. In the case 
of the soda pulp, only water had a beneficial effect. Weak bleach 
solution, sodium hydroxide (2 per cent), and hydrochloric acid 
(2 per cent) produced a loss of strength. With the sulphite 
pulp, bleach solution and hydrochloric acid during the first week 
produced an increase in strength equal to that obtained with 
water, but further soaking caused a marked decrease. Calcium 
chloride (2 per cent), sodium chloride (5 per cent), and zinc 
chloride (1 per cent) caused decided increases in strength. 

In mucilage formation, according to Minor, 1 the first step is 
the production of soluble mucilaginous dextrins that are adsorbed 
by the pure cellulose to form the insoluble aggregate called 
hydrocellulose. Treatment with hot water so hastens hydrol- 
ysis and weakens adsorption that the mucilaginous products 
are destroyed more rapidly than formed. When triturated and 
non-triturated samples of parchment pulp were washed with 
water, the triturated sample was decidedly more alkaline to 
methyl red, and when washed with hot neutral water, the alka- 
linity increased. On washing later with cold water both pulps 
became acid to methyl red. Washing with hot water again 
developed alkalinity. This alternation of basicity and acidity 
could be continued for some time, indicating that acids are not 
produced by beating in the absence of an oxidizing agent. The 
theory is advanced that hydration develops more rapidly with 
hot water than it can be washed out, thus leaving the pulp 
alkaline; with cold water the reverse is true. 

The addition of basic dyes to beaten wood pulps lowers their 
copper number. 2 

Prolonged digestion of gelatinized pulps with hot water reduces 
the hygroscopicity and copper number. 3 Long washing with 
cold water has the same effect, though not to the same extent. 
The pulps, in the table on page 236, before beating were given 
an acid treatment in accordance with G. P. 303498. 

1 J. E. Minor, J. Ind. Eng. Chem., 13 (1921), 131. 

2 F. B. Siebert and J. E. Minor, Paper 24 (1919), 1007. 

3 C. G. Schwalbe and E. Becker, Z. angew. Chem., 33 (1920), 58-59. 



236 



CHEMISTRY OF CELLULOSE AND WOOD 





Original mucilage 


Mucilage digested 

with hot water 

24 hours 


Washed mucilage 


Mucilage from 
hydrolyzed 


Copper 

number 

corrected 


Water content 1 


Copper 

number 

corrected 


Water 
con- 
tent, 1 
per cent 


Copper 

number 

corrected 






Dried at 
room 

temper- 
ature, 

per cent 


Dried at 

120°, 
per cent 


Water 
con- 
tent, 1 
per cent 


Soda pulp 

Sulphite pulp . . . 


11.35 

8.59 


37.85 
32.65 


25.16 
22.95 


3.95 
1.03 


16.64 
15.11 


8.59 
1.69 


26.45 
22.65 



1 Water content in air saturated with water vapor. 

It will be noted from the above table that the gelatinized 
pulps have a greatly increased hygroscopicity and that this is 
decreased by drying at 120°. Parchment paper (pergamyn) 
showed a reduction in hygroscopicity of 24.01 to 14.93 per cent 
after heating at this temperature for 2 hours. The pulp in the 
form of paper required an exposure of 3 to 4 weeks to water vapor 
to reach maximum absorption. Ordinary cellulose can be satis- 
fied by drawing a stream of moist air over it for 5 hours. Parch- 
ment papers showing a hygroscopicity of 17.51 to 23.23 per cent 
had this value increased to 30.42 to 32.76 per cent by further 
beating. 

Pulps having a high copper number, i.e., retaining considerable 
quantities of lignin, hemicellulose, hydrocellulose, or oxycellu- 
lose, are particularly suitable for gelatinizing. It is for this 
reason that Mitscherlich sulphite pulp lends itself so well to the 
manufacture of pergamyn papers. On account of the low tem- 
perature at which this pulp is cooked, a large part of the mucilage- 
forming materials are retained. 1 

The ease of gelatinization of pulps varies with the species 
of wood, the degree and method of cooking, and the degree of 
bleaching. In general, for unbleached pulps, the amount of 
beating required increases in the order: Mitscherlich sulphite, 
kraft sulphate, Ritter-Kellner sulphite, soda, cotton, and 

1 C. G. Schwalbe and E. Becker, Z. angew. Chem., 33 (1920), 57; C. G. 
Schwalbe, Wochbl. Papierfabr , 51 (1920), 1486; F. B. Siebert and J. E. 
Minor, Pulp Paper Mag. Can., 18 (1920), 939. 



GELATINIZED CELLULOSE 237 

mechanical wood pulp. A rough measure of ability to gelatinize 
may be obtained by the rattle or " tinny' ' character of the paper 
made from soda or sulphate pulps that have been cooked and 
beaten to the same extent. By this criterion ease of gelatiniza- 
tion increases in the order of the following groups of Gymno- 
spermai: (1) Picea; (2) Pinus; (3) Pseudotsuga (Douglas fir); (4) 
Thuja, Cupressus, Chamcecy parts, Juniperus; and (5) Sequoia 
(redwood). 

The fibers of various species of Prunus, obtained by maceration 
of the wood in a mixture of strong nitric acid and potassium 
chlorate, were markedly swollen. This was not the case with 
various species of conifers. 1 

Action of Bacteria. — It is a common observation that celluloses 
gelatinized by beating undergo fermentation after standing 
some time, eventually losing their adhesive properties. In 
turn, a certain amount of bacterial action is beneficial. The 
fibers of Hedychium coronarium, after beating in a Hollander 
and storing in the wet condition, became more slimy and gelati- 
nous on aging. 2 At the end of the fifth week it gave a paper having 
the maximum bursting strength. After the fourth week putre- 
faction set in, and after the fifth week gradually destroyed the 
material giving the paper increased strength, thus undoing the 
original beneficial effects of aging. 

When finely divided bleached cellulose is stored under water 
for some time, fermentation produces a mucilaginous material. 3 
The latter under suitable conditions becomes sufficiently large 
to permit manufacture of a paper without the addition of sizing 
substances. The paper on drying at 120° becomes water resist- 
ant, due to the irreversible state of the mucilage. 

The degree to which pulp has been gelatinized by beating may 
be judged by a determination of its adsorptive capacity for 
sodium hydroxide in alcoholic solution, 4 and for substantive 
dyes, such as diamine blue and methylene blue. 3 The hygro- 
scopic moisture method 5 gives good results with a humidity of 

1 H. S. Conrad and W. A. Thomas, Proc. Iowa Acad. Sci., 26 (1919), 333- 

2 C. Beadle and H. P. Stevens, J. Soc. Chem. Ind., 32 (1913), 217-218. 
3 0. Ruff, G. P. 311772 (1918). 

4 J. F. Briggs, Chem. Ztg., 34 (1910), 455. 

& C. G. Schwalbe, Wochbl. Papierfabr., 54 (1923), 22. 



238 CHEMISTRY OF CELLULOSE AND WOOD 

80 to 90 per cent. The humidity can be controlled by exposing 
the pulp to sulphuric acid of a definite concentration. The 
method is slow but accurate except when the humidity reaches 
the dew point. 

Much ingenuity has been expended on mechanical devices to 
secure a rapid determination of gelatinization. Viscosity 
methods based on a falling sphere, flow through an orifice, or 
torsion, 1 are not dependable, owing to the non-homogeneity of 
the pulp. The apparatus in general use are based directly or 
indirectly on the rate of flow of water through a screen using a 
definite amount of pulp at a certain dilution. 2 They are gen- 
erally known as sedimentation testers. The Schopper-Riegler 3 
apparatus as modified by Green 4 is a very useful type. 

Sedimentation tests are of questionable value unless made on 
the same stock. 5 Widely different chemical pulps may give 
identical draining values; further, the same pulp beaten under 
apparently the same conditions may give sedimentation curves 
varying widely. This variation, particularly characteristic of 
sulphite pulp, is probably due to lack of uniform cooking 
conditions. 

Gelatinization of Lignocellulose. — The capability of ligno- 
cellulose to gelatinize was formerly considered to be slight. 6 
According to de Cew, 7 wood does not gel as readily by mechanical 
treatment in the cold as does cellulose, but unlike cellulose its 
gelation is promoted by heat. Gelatinization has been obtained 
by the viscose reaction, 8 and by the limited action of strong 
mineral acids. Schwalbe 9 gelatinizes mechanical pulp by beating 

1 E. W. L. Skark, Papier-Fabr., Fest u. Auslandsheft, 12 (1914), 87. 

2 P. Klemm, Wochbl. Papierfabr., 39 (1908), 1215; E. W. L. Skark, 
Papier-Fabr., 11 (1913), 1358, 1381, 1417; 19 (1921), 569; A. Schopper, 
Papier-Ztg., 39 (1913), 642; "Snowshoe," Pulp Paper Mag. Can., 16 (1918), 
793. 

3 M. Riegler, U. S. P. 1193613 (1916). 

4 A. B. Greex, Paper 17, 24 (1915), 21. 

5 O. Kress and G. C. McXaughtox, Paper 20, 17 (1917), 13-17. 

6 C. F. Cross and E. J. Beyax, "Paper Making" (1900), p. 169; R. W. 
Sixdall, "An Elementary Manual of Paper Technology" (1906), p. 52. 

7 J. A. de Cew, J. Soc. Chem. Ind., 36 (1917), 357. 

8 P. B. Lach, E. P. 12324 (1912); Portolac Holzmasse Ges., Austrian P. 
64798 (1914); J. A. de Cew, U. S. P. 1140799 (1915). 

9 C. G. Schwalbe, G. P. 303305 (1917). 



GELATINIZED CELLULOSE 239 

it in a Hollander in the presence of an oxidizing agent such as 
nitric acid. 

The use of the Plauson colloid mill for gelatinizing cellulose 
and lignocellulose has been described, and alkali mentioned as a 
peptizing agent. 1 Three 15-liter mills in series running at 9000 
r.p.m. will in 1 hour reduce 200 kilograms of cellulose to such 
fineness that 85 per cent of the particles will be about 0.001 
millimeter in diameter. 2 The power consumption, 15 to 20 
kilowatts per kilogram of substance, is very high. 3 

It was found that by grinding wood in water in a ball mill for 
24 to 48 hours the conifers could be readily gelatinized, but the 
hardwoods showed little change; however, the addition of suffi- 
cient sodium hydroxide to form a 0.25 per cent solution had a 
marked peptizing action, particularly on the hardwoods. 4 The 
reaction has been applied to the bark of trees and the cereal 
straws. The higher the hemicellulose content of a substance, the 
more readily it gelatinizes. With iodine solution, gelatinized 
wood stains blue, like cellulose. In the dry condition it forms a 
dense, horn-like mass having a specific gravity of 1.5. When 
dry wood was ground in a solution of sodium hydroxide in 
absolute methyl alcohol, then filtered and washed with alcohol, 
the residue on drying formed an easily friable mass that remained 
unaffected by water; however, when water was used for washing 
the wood while still wet with alcohol, it became gelatinous at 
once. This indicates that gelatinization takes place in two 
stages: the wood first acquires by mechanical comminution a 
peculiar affinity for water with which it then combines to form a 
gelatinous product. 

It does not appear to be possible to produce gelatinization by 
either mechanical or chemical methods without destroying the 
regular orientation of the crystallites or micellae about the axis of 
the fiber. Straight pressures of 500,000 pounds to the square 
inch applied to wood and chemical pulp in either the air-dry 

1 B. Block, Z. angew. Chem., 34 (1921), 28; cf. H. Plauson, Chem. Ztg., 
44 (1920), 535, 565; Z. angew. Chem., 34 (1921), 469, 473; A. W. Kenney, 
Chem. Met. Eng., 27 (1922), 1080. 

2 A. Forster and J. Reilley, J. Soc. Chem. Ind., 41 (1922), 438R. 

3 W. J. Kelley, J. Ind. Eng. Chem., 15 (1923), 926. 

4 A. W. Schorger, J. Ind. Eng. Chem., 15 (1923), 812. 



240 CHEMISTRY OF CELLULOSE AND WOOD 

condition, or wet with water or dilute sodium hydroxide, did not 
produce gelatinization. 1 Microscopic examination indicated no 
visible disintegration of the fibers. More promising results are 
obtained with calenders where there is a certain shearing action. 
The effect of mechanical disintegration on cellulose is a field 
worthy of further investigation. Either dry or wet beating 
results in a decrease in the viscosity of the cuprammonium solu- 
tions of the cellulose so treated. 2 The mechanical disintegration 
of cellulose by beetling results in a chalky powder that no longer 
stains blue with iodine solution. 3 Tests for oxy cellulose and 
hydrocellulose were negative. Treatment with 17.5 per cent 
sodium hydroxide showed 44.2 per cent a-cellulose, 48.6 per cent 
/3-cellulose, and 7.2 per cent 7-cellulose. 4 The yield of acetate on 
acetylation was 122 per cent in comparison with 178 per cent for 
normal cellulose. When heated in water above 77°, the suspended 
cellulose underwent an apparent hydration effect, the powder 
changing to a flocculent curd. The phenomenon was reversible, a 
powder being obtained on cooling. 

Cellulose Solvents. Cuprammonium Solution. — There is no 
true solvent for cellulose. The various so-called solvents per- 
mit the recovery of a modified cellulose only. On dilution of the 
solvent, gelatinous cellulose usually precipitates. The extent to 
which the cellulose is degraded depends upon the nature of the 
solvent and the length of time that the cellulose remains in the 
solvent. 

Schweizer's reagent, 5 an ammonical solution of cupric oxide, 

1 Unpublished results. 

2 P. Waentig, Text. Forsch., 3 (1921), 154. 

3 M. Fort, J. Soc. Dyers Colourists, 34 (1918), 9-10. 

4 C. F. Cross and E. J. Bevan, J. Soc. Dyers Colourists, 34 (1918), 
215-218. 

5 E. Schweizer, J. prakt. Chem., 72 (1857), 109; 78 (1859), 370. 

The discovery of the solvent action of cuprammonium solution is invari- 
ably attributed to Schweizer, usually incorrectly spelled Schweitzer in the 
literature. E. A. Parnell, in "Life and Labours of John Mercer," London 
(1886), p. 214, states: "Mercer appears to have been the first to notice the 
remarkable action which the ammoniacal solution of cupric oxide exerts on 
cotton fibre. He found that cotton is completely soluble in the liquid 
obtained by dissolving hydrated cupric oxide to saturation in ammonia of 
sp. gr. 0.920 and diluting with three measures of water. On neutralizing 
this solution with sulphuric acid, the organic matter was precipitated, and 



GELATINIZED CELLULOSE 241 

dissolves cellulose rapidly. Its solvent power is very limited. 1 
After treatment with an excess of cotton for a few minutes, the 
cuprammonium solution not only loses its ability to dissolve more 
cotton, but fresh cotton is not even swelled by it. The solvent 
power is greatly reduced or prohibited by the presence of salts. 2 
Ammoniacal nickel oxide does not attack cellulose. 3 
Gibson, 4 following Dawson's 5 method for preparing crystalline 
cupric hydroxide, prepares the cuprammonium solution as 
follows: 60 grams of copper sulphate are dissolved in 1 liter of 
hot water and a few drops of sulphuric acid added. The solu- 
tion is cooled to 50° and ammonia of sp. gr. 0.880 is added until 
the precipitation of basic copper sulphate is complete, any excess 
of ammonia being neutralized with a few drops of sulphuric acid. 
After settling and decanting, the precipitate is washed with hot 
water by decantation, 200 cubic centimeters of 20 per cent 
sodium hydroxide added, and the whole shaken at the ordinary 
temperature. The precipitate is washed with cold water by 
decantation until it is free from alkali and salts. It is then 
collected on a porous plate and dried at 40°. The dried product 
is transferred to an aspirator bottle, 800 cubic centimeters of 
ammonia water containing 200 to 210 grams of ammonia per 
liter added, and the whole well shaken. The excess cupric 
hydroxide is allowed to settle. The solution is then filtered 
through glass wool. When used for viscosity measurements, 
the amount of copper present is determined and sufficient 
ammonia is added to give a solution containing 11 grams of 
copper and 200 to 210 grams of ammonia per liter. 



the latter became free from copper by an excess of sulphuric acid." Mercer 
recognized that a low temperature and absence of salts increased the sol- 
vent power of the solution. A solution prepared in accordance with the 
above would have a low solvent power. Why he diluted with water is not 
apparent. 

The date of these discoveries is not stated. Mercer was born Feb. 21, 
1791 and died Nov. 30, 1866. 

1 C. Cramer, J. prakt. Chem., 73 (1858), 7. 

2 J. Schlossberger, Ann., 107 (1858), 23. 

3 J. Schlossberger, I.e., 21. 

4 W. H. Gibson, J. Chem. Soc., 117 (1920), 492. 
s H. M. Dawson, J. Chem. Soc., 95 (1909), 371. 



242 CHEMISTRY OF CELLULOSE AND WOOD 

The cuprammonium solution prepared from basic copper 
sulphate by Ost's 1 method contains tetrammine copper sulphate. 2 

Joyner 3 prepared the reagent by bubbling air through a 
mixture of copper turnings in strong ammonia, a little sucrose 
being added to hasten solution of the copper. In this way it is 
possible to prepare solutions containing more than 30 grams of 
copper per liter. This method has been used industrially. 4 A 
solution containing 40 to 50 grams of copper per liter can be 
prepared by means of compressed air and low temperatures. 5 

Mulder 6 prepared solutions of cellulose by repassing ammonia 
through alternate layers of copper turnings and cotton in a 
separatory funnel. 

It has been stated 7 that an ammoniacal solution of cuprous 
chloride dissolves cellulose rapidly; in the absence of oxygen, 
however, no solution takes place. 8 A solution prepared by 
treating 90 grams of copper oxychloride with 850 to 900 cubic 
centimeters of ammonia water of sp. gr. 0.93 will dissolve 100 
grams of cellulose. 9 In this case, the presence of chlorides is 
assumed to be beneficial. 

Primary aliphatic diamines saturated with cupric hydroxide 
have the property of dissolving cellulose to a marked extent, 
being superior in this respect to equal concentrations of 
ammonia. 10 With ethylene diamine there is formed the com- 
pound Cu(C2H 8 N 2 )2.(OH)2. Solutions containing as little as 
2 to 8 per cent of ethylene diamine are effective as solvents. 
Cellulose is only slightly soluble in corresponding aqueous solutions 
of primary aliphatic amines and insoluble in secondary. 



1 H. Ost, Z. angew. Chem., 24 (1911), 1893. 

2 W. H. Gibson, I.e. 

3 R. A. Joyner, J. Chem. Soc, 121 (1922), 1511. 

4 E. Bronnert, M. Fremery, and J. Urban, G. P. 115989 (1900). 

5 C. Suvern, "Die Kuntsliche Seide," p. 100. 

6 E. Mulder, Jahresber. Chemie, (1863), 566. 

7 M. Rosenfeld, Ber., 12 (1879), 956. 

8 C. F. Cross and E. J. Bevan, "Cellulose" (1895), p. 246. 

9 R. Pawlikowski, F. P. 403488 (1909). 

io W. Traube, Ber., 44 (1911), 3319; G. P. 245575 (1911); E. Friedrich, 
U. S. P. 850571 (1907). 



GELATINIZED CELLULOSE 243 

The solvent power of cuprammonium solutions is greatly 
increased by the presence of sodium hydroxide. 1 The solubility 
of cupric hydroxide in the caustic alkalis alone is slight. One 
liter of sodium hydroxide of sp. gr. 1.345 to 1.370, or potassium 
hydroxide of sp. gr. 1.453 to 1.498, will dissolve about 7.8 grams 
of copper hydroxide. 2 

When cellulose is treated with a fixed alkali, an alcoholate 
is apparently formed. This reacts with cupric hydroxide to 
form an alkali-copper-alcoholate. 3 The fixed alkali may be 
replaced by metallic ammonium hydroxides. Cellulose forms 
with Schweizer's reagent an alcoholate, the hydrogen of the free 
hydroxyl groups being replaced by copper to give a salt as 
follows : 

[Cu(NH 3 ) 4 ](OH) 2 + [HO-C 6 H 7 2 (OH) 2 ] 2 > 

Cellulose 

[Cu(NH 3 )4]-[0-C 6 H 7 2 (OH) 2 ] 2 + 2H 2 
(A) 

(A) + Cu(OH) 2 — ► [Cu(NH 3 )4]-[0-C 6 H 7 2 (OH)0] 2 Cu + 2H 2 0. 

According to Hess, 4 the compound [Cu(NH 3 ) 4 HC 6 H 7 05Cul2 is 
formed. 

Caustic soda coagulates a cuprammonium solution of cel- 
lulose, forming a transparent, blue copper-alkali compound 5 
of the composition Na 2 (Ci 2 Hi 6 Oi )Cu. 

Wood cellulose was dissolved in a solution of copper carbonate 
in ammonia. On heating this solution there was deposited a 
copper-cellulose compound considered 6 as having the definite 
composition 11CuO-2C 6 Hi O 5 -2H 2 O. 

Metallic zinc gradually removes all the copper from cupram- 
monium solutions of cellulose, giving a colorless liquid having 

1 M. Prud'homme, F. P. 344138; P. Friedrich, F. P. 410882 (1909). 

2 E. Justin-Mueller, Compt. rend., 167 (1918), 779. 

3 W. Traube, Ber., 54B (1921), 3220; 55B (1922), 1899; 56B (1923), 268. 

4 K. Hess et al., Ann., 435 (1923), 1. 

5 W. Normann, Chem. Ztg., 30 (1906), 584; K. Hess et al, Ber., 55B. 
(1922), 2432; 56B (1923), 587. 

6 H. Riesenfeld and F. Taurke, Ber., 38 (1905), 2798. 



244 CHEMISTRY OF CELLULOSE AND WOOD 

properties similar to the copper compound. 1 Some cellulose 
usually separates during the treatment. 

The keeping quality of cuprammonium solutions of cellulose 
is improved by the addition of mannitol and other polyatomic 
alcohols. 2 

Sodium chloride, and various other salts, precipitate cellulose 
from the solution. The precipitated cellulose is colored violet 
to wine red by tincture of iodine. 3 Besides salts, the cellulose 
is precipitated by acids, even carbonic acid, and sugar. 4 

Strong ammonia added to copper sulphate forms tetrammine 
copper hydroxide, 5 Cu(NH 3 ) 4 .(OH) 2 . The ions Cu n NH 3 " and 
Cu" are also present. 6 

In general, the higher the percentage of copper in Schweizer's 
reagent the greater is its solvent power. Grimaux, 7 as a result 
of experiments on dialysis, concluded that it was the non- 
dialyzable portion that acted as a solvent. According to 
Connerade, 8 the solubility of cellulose is proportional to the 
concentration of the colloidal cuprammonium hydroxide, the 
concentration being greater for solutions prepared at low tem- 
peratures. The fixation of copper follows the adsorption laws, 
though there are no stoichiometrical relations between the weight 
of cuprammonium hydroxide and the weight of the cellulose. 
This is due to solution of cellulose being a colloidal phenomenon. 9 
The colloidal portion of the cuprammonium hydroxide combines 
with the cellulose to give an adsorption compound soluble in 
ammonia. Cuprammonium solutions have almost the same 
tension of ammonia as solutions of ammonia of equal concentra- 
tion; hence colloidal copper hydroxide must be present, though 
it plays a secondary role in the solution of the cellulose. The 
presence of cellulose in solution does not change the ammonia 

1 E. Mulder, Jahresber. Chemie, (1863), 566. 

2 P. Friedrich, F. P. 400321 (1909). 

3 J. Schlossberger, Ann., 110 (1859), 247. 

4 E. Mulder, Jahresber. Chemie. (1863), 566. 

5 J. Locke and J. Forssall, Am. Chem. J., 31 (1904), 268. 

6 H. M. Dawson, J. Chem. Soc., 89 (1906), 1666. 

7 E. Grimaux, Compt. rend., 98 (1884), 1434. 

8 E. Connerade, Bull. soc. chim. Belg., 28 (1914), 176; cf. Soc. Anon. 
Crinoid F. P. 401741 (1908); U. S. P. 947715 (1910). 

9 E. Berl and A. G. Innes, Z. angew. Chem., 23 (1910), 987. 



GELATINIZED CELLULOSE 



245 



tension. Dialysis gives a product containing 69.5 per cent cellu- 
lose and 30.5 per cent copper hydroxide. 

The solubility of cellulose in a cupram- 
monium solution depends not only on 
the solution itself but on the condition of 
the cellulose. Cellulose dissolves almost 
immediately if it has been mercerized 
with caustic soda. 1 Obviously, the cel- 
lulose should be free from fats and waxes. 
The cuticle of cotton is insoluble and 
offers mechanical resistance (Fig. 6) to 
the solution of the fiber; hence the use 
of bleaching agents, such as calcium 
hypochlorite, that destroy the cuticle, 
greatly facilitates solution. 2 

Statements regarding the solubility of 
cellulose accordingly vary greatly. A 
solution containing 150 to 200 grams of 
ammonia and 18 grams of copper per 
liter, prepared by drawing air through 
ammonia in contact with copper, dis- 
solved 3 per cent of cellulose. The 
solubility of cellulose based on the cop- 
per present is stated to be as follows: 




Fig. 6. — Cotton swollen 
in cuprammonium solution. 
cf, shreds of cuticle; cr, ring 
of cuticle; ce, distended cell- 
ulose; i, dried protoplasmic 
lining of the banded lumen, 
greatly shrunken and hence 
wrinkled transversely. 

(After von Hohnel.) 



Cellulose Dis- 
solved by 1 Part 
of Copper 



2 R. Pawlikowski, F. P. 403488 (1909). 

2.5 P. Friedrich, F. P. 410882 (1909). 

3 H. Pauly, G. P. 98642 (1897). 

2.3 R. A. Joyner, /. Chem. Soc., 121 (1922), 1511. 

According to Pauly, the solution contained 45 grams of cellu- 
lose per liter, or approximately 4.5 per cent. Fremery and 
Urban 3 obtained a 4 per cent solution with ordinary cotton. 

Cellulose dissolved in cuprammonium solution is used in the 
manufacture of Pauly silk (GlanzstofT ) . The solution should 
contain at least 5 per cent of cellulose, preferably 8 to 10 per 



1 E. Thiele, Rev. prod, chim., 3 (1900), 325. 

2 M. Fremery and J. Urban, G. P. 111313 (1899). 

3 M. Fremery and J. Urban, I.e. 



246 CHEMISTRY OF CELLULOSE AND WOOD 

cent. 1 To obtain solutions of these concentrations, it is neces- 
sary to use a cellulose that has been mercerized or slightly 
oxidized by a carefully regulated treatment with calcium hypo- 
chlorite. The solution, after filtering, is spun through capillary 
orifices at a pressure of 2 to 4 atmospheres. The coagulating 
bath is a 45 to 50 per cent solution of suphuric acid or a 30 to 
40 per cent solution of sodium hydroxide. The manufacture of 
silk by the cuprammonium process is expensive but is still 
carried on, as the product has a somewhat greater resistance to 
water than the other kinds of artificial silks. 

Prud'homme 2 states that cuprammonium solution has an 
oxidizing action on cellulose similar to hydrogen peroxide; 
hence, contrary to the general opinion, cellulose is attacked 
by the solvent. The ammonia likewise undergoes oxidation 
with the formation of nitrites. 

Cellulose cuprammonium solutions are of slight stability. 
Precautions should be taken to guard them from air and light. 
A small amount of oxygen has an enormous effect on the viscos- 
ity. Joyner 3 believes that there are two kinds of cellulose: 
"A" having a maximum viscosity of 57,000 seconds for a 2 per 
cent solution at 20°; "B" having the low viscosity of a second for 
a 5 per cent solution. Sixteen grams of oxygen will react with 
about 2,100 grams of cellulose, converting all of "A" into "B." 
When cellulose, in a caustic soda solution, was treated with 1.5 
cubic centimeters of sodium hypochlorite per gram, the viscosity 
of a 4 per cent solution in cuprammonium was reduced to 4 
seconds. 

The viscosity of the solution affords a useful means, when 
stringent precautions are observed, 4 of determining if the cellu- 
lose is normal or degraded. 5 Joyner 6 recommends for general 
use, a solution of 20 grams of cellulose in a reagent containing 13 
grams of copper and 200 grams of ammonia per liter. In the 

1 Ullmann, "Enzyklop. tech. Chemie," 7 (1919), 330. 

2 M. Prud'homme, Compt. rend., 112 (1891), 1374. 

3 R. A. Joyner, /. Chem. Soc, 121 (1922), 2395. 

« W. H. Gibson, /. Chem. Soc., 117 (1920), 479; M. Nakano, J. Chem. 
Ind. (Japan), 25 (1922), 899; R. A. Joyner, I.e. 

5 C. Piest, Z. angew. Chem., 25 (1912), 2518; H. Ost, Ibid., 24 (1911), 
1892. 

6 R. A. Joyner, J. Chem. Soc., 121 (1922), 1511. 



GELATINIZED CELLULOSE 



247 



case of a cellulose of low viscosity, 40 grams of the product are 
dissolved in a reagent containing 20 grams of copper and 200 
grams of ammonia per liter. 

Changes in the viscosity of the solution, attributed to depoly- 
merization of the cellulose molecule, show that the solvent is not 
inert and prevents sharp differentiation between celluloses. 
Ost 1 obtained the highest viscosity for a nitration cotton that 
had been boiled with dilute caustic soda. The lowest viscosities 
and easiest solubilities were shown by filter paper and bleached 
straw cellulose. Sulphite pulp purified for celluloid manufacture 
showed a viscosity of 10.7°. The viscosity of cotton was reduced 
from 30.2 to 13.6° by a previous heating at 120 to 125° for 15 
hours. Oxycellulose and hydrocellulose show a low viscosity. 
The residues from heating these degraded celluloses with alkalis 
is usually considered as pure cellulose; yet they still show low 
viscosities. 

Nakano 2 found the following changes in the viscosity of 
cellulose (S. S. filter paper No. 590) after heating for 4 hours at 
various temperatures: 



Temperature . 
Viscosity .... 



Control 
2.26 



100-105 c 
2.20 



120-125° 
2.14 



140-145° 
1.87 



160-165° 
1.62 



180-185° 
1.51 



Cotton, 10 grams, mercerized with 200 cubic centimeters of 
sodium hydroxide solutions for 24 hours, showed the following 
viscosities : 



NaOH, per cent. 
Viscosity , 



Control 
4.22 



11 
3.59 



17 
3.03 



20 
2.01 



Depolymerization of the molecular cellulose aggregate, other 
conditions being equal, is due to light. 

Crystallization of Cellulose. — In 1893, Gilson 3 reported that 
cellulose could be obtained in the form of crystals. Cotton and 
thin sections of plants were treated with Schweizer's reagent. 
After becoming strongly swollen, the cellulose was treated with 

1 H. Ost, I.e. 

2 M. Nakano, I.e. 

3 E. Gilson, La Cellule, 9 (1893), 397-440. 



248 CHEMISTRY OF CELLULOSE AND WOOD 

ammonia until the preparation was almost colorless. This was 
followed by water, and dilate hydrochloric acid or acetic acid. 
With proper precautions the cellulose was converted into a mass 
of isotropic spherocrystals. Crystals were obtained from the 
wood of two species of gymnosperms and from the celluloses 
from various species. The crystals were insoluble in boiling 
2 per cent sulphuric acid and, since they were without action 
on polarized light, they are to be considered as crystallites rather 
than crystals. 

Gilson's findings were confirmed by Johnson 1 and Butchli. 2 
De Mosenthal 3 identified the crystals microchemically as cuprous 
chloride. He states that when acetic acid is used no crystals 
are obtained. 

According to the more recent investigations of Alexandrowicz, 4 
spherocrystals of cellulose can be obtained in a few hours by 
immersing in concentrated ammonia a capillary tube containing 
a dilute solution of cellulose in cuprammonia. The crystals 
lose their structure on complete drying. While the cellulose of 
plants is strongly doubly refractive, the cellulose crystals are 
isotropic or very weakly anisotropic. Hemicelluloses increase 
the optical anisitropy of cell tissues. 

Salts as Cellulose Solvents. — A sheet of paper plunged into 
a hot, concentrated solution of zinc chloride swells and dissolves 
without coloration. 5 Before undergoing saccharification the 
cellulose passes through a stage in which it is stained blue by 
iodine. When cotton is moistened with about 10 parts of 40 
per cent zinc chloride solution, and the mixture gradually warmed 
to the temperature of a hot water bath, a syrupy solution is finally 
obtained. 6 Water precipitates gelatinized cellulose containing 
18 to 25 per cent of zinc oxide that can be removed by treatment 
with dilute hydrochloric acid and washing. 

The solutions of cellulose obtained with zinc salts have 
been used for producing threads and filaments, but its use for 

1 D. S. Johnson, Botan. Gaz., 20 (1895), 16-22. 

2 O. Butchli, " Untersuchungen iiber Strukturen," Leipzig (1898), 
p. 200. 

3 H. de Mosenthal, J. Soc. Chem. Ind., 23 (1904), 293. 

4 J. S. Alexandrowicz, Arch. ges. Physiol., 150 (1913), 57-86. 

5 C. Barreswil, J. pharm. chim., 21 (1852), 205. 
e C. F. Cross and E. J. Bevan, "Cellulose," p. 8. 



GELATINIZED CELLULOSE 249 

these purposes has practically disappeared, as the solvent has a 
destructive action on the cellulose. The solvents consisted of 
zinc chloride, bromide, or iodide; or bismuth chloride or bromide 
separately or as mixtures; 1 and zinc chloride with 5 per cent of a 
suitable alkaline earth chloride such as calcium chloride. 2 As 
with other solvents, the cotton is dissolved more readily and gives 
more concentrated solutions if previously mercerized or converted 
to hydrocellulose. 

It has been observed by Deming 3 that numerous metallic 
salts having an acid reaction in aqueous solution are capable of 
dissolving cellulose. A pyridine or quinoline solution of zinc 
chloride will not dissolve normal cellulose even on heating. Their 
solvent power is greatly increased by the addition of strong 
hydrochloric acid. Four parts of stannous chloride dissolved 
in one part of water will dissolve cellulose at 100°, while the 
solution of the salt in concentrated hydrochloric acid will dissolve 
it in the cold. 

When dissolved in strong hydrochloric acid, the solvent powers 
of the various salts were: 



Solvent power 



Salt 



Good 

Moderate . 

Poor 

None 



SbCl 3 , SbCl 5 , HgCl 2 , BiCl 3 , SnCl 4 , TiCl 4 
CoCl 2 , AuCl 3 , U0 2 C1 2 , CeCl 3 , CrCl 3 
CdCl 2 , Cu 2 Cl 2 , T1C1 3 , VC1 4 
MnCl 2 , CuCl 2 , NiCl 2 , AgCl, PbCl 2 , T1C1, 
AICI3, ThCl^ AsCl 3 



Solutions of chlorides of the alkalis and alkaline earths in hydro- 
chloric acid do not dissolve cellulose, owing probably to their 
slight solubility in this medium; however, solutions of CaCl 2 , 
CaBr 2 , BaCl 2 , MgBr 2 , LiCl, and KBr in formic acid, alone or 
admixed with hydrochloric acid, do dissolve cellulose. 

Ferric chloride, FeCl 3 .6H 2 0, in the molten condition dissolves 
cellulose rapidly. 

1 J. Wynne and L. Powell, E. P. 16805 (1884). 

2 W. P. Dreaper and H. Tompkins, E. P. 17901 (1897). 

3 H. G. Deming, J. Am. Chem. Soc., 33 (1911), 1515-1525. 



250 CHEMISTRY OF CELLULOSE AND WOOD 

In 1905, Dubosc 1 accidentally discovered that some thiocyante 
solutions, especially when heated, were cellulose solvents giving 
solutions adapted to spinning. Solution was attributed to the 
formation of a cellulose thiocyanic ester. Dubosc used the 
ammonium, potassium, calcium, and iron salts. Potassium thio- 
cyanate 2 melted with a few drops of water will dissolve cellulose 
at 200°. Any soluble thiocyanate will also dissolve alkali- 
cellulose. 3 

The action of the thiocyanates has been carefully examined by 
Williams. 4 It was found that, in the case of a neutral salt, there 
is a definite connection between, the boiling point, viscosity, 
and heat of dilution of its solution and its ability to dissolve 
cellulose. These properties must be brought within definite 
limits. None of the thiocyanate solutions dissolved cellulose 
until the boiling point of the solution reached 133 to 134°. 
There must be a positive heat of dilution not exceeding about 
3500 calories, and the viscosity must be at least 3.3 times that of 
water at 20°. 

Lithium thiocyanate does not dissolve cellulose until its solu- 
tion is concentrated to a boiling point of 165°, where the proper 
viscosity is reached; but when the viscosity is increased by the 
addition of compounds that increase the viscosity without dimin- 
ishing the heat of dilution, a solvent can be obtained having a 
boiling point 30° lower. The addition need not be a thiocyanate 
for thiourea or dicyandiamide may be used. In this way, thio- 
cyanates not in themselves solvents can be made active by the 
proper addition of other thiocyanates or compounds. 

Calcium thiocyanate is the best solvent. A solution boiling 
at 133° will dissolve wood pulp by heating to 90°. The solution 
must be neutral or acid; if alkaline, no solution takes place. 
A little acetic acid assists solution. The addition of 1 volume 
of calcium chloride solution, b.p. 133°, to 2 volumes of calcium 
thiocyanate solution, b.p. 133°, does not decrease the solvent 

1 A. Dubosc, Bull, Rouen, 33 (1905), 318; Rev. prod, chim., 26 (1923), 507. 

2 H. G„ Deming, I.e., 1523. 

3 A. Dubosc, Bull. Rouen, 36 (1908), 272. 

4 H. E. Williams, J. Soc. Chem. Ind., 40 (1921); 221T; Mem. Proc. 
Manchester Lit. Phil. Soc., 65(1921), 12; cf. U.S.P. 1301652; E.P. 123784 
(1917). 



GELATINIZED CELLULOSE 251 

power of the latter. With solutions of b.p. 140°, equal volumes 
of the two solutions may be used. One hundred cubic centi- 
meters of the mixture will dissolve 7 grams of cotton or wood pulp. 
Unlike sulphuric acid, this solution can be used for parchmentiz- 
ing paper containing lignocellulose. It can be used for making 
vulcanized fiber and in spinning threads, coagulation taking 
place in 50 to 90 per cent alcohol. 

While it might be assumed that cellulose would undergo but 
slight degradation in a neutral solvent, this does not appear to 
be the case. Beck 1 dissolved cellulose in a concentrated solution 
of calcium thiocyanate at 120°. The solution set to a clear jelly 
on cooling. Films prepared from the solution had less strength 
than for cellulose prepared from viscose or a cuprammonium 
solution. It was found that the strength of the films could be 
increased 80 per cent in both the wet and dry condition by 
immersion in a concentrated solution of calcium thiocyanate at 
room temperature for 24 hours. The strength of the film was 
greater also if the calcium thiocyanate was not completely 
removed by washing. 

An interesting theory of the solvent action of neutral salt 
solutions on cellulose has been given by Williams. 2 The hydroxyl 
groups of the cellulose unit combine with the salt complex in 
place of water molecules, behaving as a substituted water group, 
thus causing the fiber to swell considerably. In this way the 
cellulose unit is brought into molecular range with the water 
molecules combined with the salt. By raising the temperature, 
the union between the salt and water molecules will weaken and 
tend to part from the parent molecule. The water thus freed 
migrates to the cellulose by which it is imbibed, causing further 
swelling of the fiber, which increases as the progressive hydration 
proceeds. The highly swollen fiber in the gelatinous condition 
then peptizes, and passes into colloidal solution. 

Octacetylcellobiose and pentacetylglucose dissolve in solu- 
tions of calcium thiocyanate, bromide, and chloride without 
saponification. 3 Cellulose nitrate can also be swollen in solu- 
tions of calcium thiocyanate or chloride without noticeable 

1 F. Beck, Z. angew. Chem., 34 (1921), 113. 

2 H. E. Williams, I.e. 

3 K. Schweiger, Z. physiol. Chem., 117 (1921), 65. 



252 CHEMISTRY OF CELLULOSE AND WOOD 

decomposition. Cellulose formate dissolves readily in a 20 
per cent solution of the thiocyanates. 1 According to Dubosc, 2 
this is not a case of simple solution but a double decomposition 
with the formation of a cellulose thiocyanic ester. 

Hydrogen Peroxide as a Cellulose Solvent. — Cellulose and its 
modifications are soluble in hydrogen peroxide of a concentration 
of 60 per cent or greater. 3 Swelling first takes place followed by 
solution. Addition of water to the viscous solution precipitates 
the cellulose in a chemically changed form. The more modified 
the original cellulose, the more readily it dissolves. Treatment 
of cellulose with alkalis increases the solubility, so that hydrogen 
peroxide is suggested as a suitable reagent for detecting 
mercerization. 

Acids as Cellulose Solvents. — Concentrated sulphuric and 
hydrochloric acids dissolve cellulose, first transforming it into a 
swollen gelatinous mass. 4 Bechamp 5 states that the hydro- 
chloric acid used had a sp. gr. of 1.20 (39.8 per cent). Pelouze 6 
observed that concentrated hydrochloric acid is an excellent 
solvent for cellulose. If the latter is precipitated with water a 
short time after solution, it resembles the cellulose obtained from 
cuprammonium solutions. After a day or two, a precipitate is 
no longer obtained and the solution strongly reduces Fehling's 
solution. 

Cellulose that has been precipitated from a cuprammonium 
solution will dissolve in hydrochloric acid of a concentration 
not effecting solution of the ordinary forms of cellulose. 7 

The solvent power of hydrochloric acid has been studied in 
detail by Willstatter and Zechmeister. 8 Acid containing 37.6 
per cent of hydrogen chloride, after contact with cellulose for 
one day, causes disintegration and gelatinization. Solution 



1 C. F. Cross and E. J. Bevan, J. Chem. Soc, 99 (1911), 1450. 

2 A. Dubosc, Caoutchouc and gutta-percha, 10 (1913), 6895. 

3 M. Bamberger and J. Nussbaum, Monatsh., 40 (1919), 411. 

4 A. Bechamp, Compt. rend., 42 (1856), 1213. 
6 A. Bechamp, Compt. rend., 100 (1885), 370. 

6 J. Pelouze, Compt. rend., 48 (1859), 327. 

7 J. Pelouze, Compt. rend., 48 (1859), 210. 

8 R. Willstatter and L. Zechmeister, Ber., 46 (1913), 2995; L. Zech- 
meister, Diss. Zurich (1913); R. Willstatter, G. P. 273800 (1913). 



GELATINIZED CELLULOSE 253 

does not take place in acid of sp. gr. 1.196 (about 39.2 per cent). 1 
The most suitable concentrations are 40 to 42 per cent. On 
shaking cellulose at room temperature with acid of 40.8 per cent, 
a clear solution is obtained in 10 seconds. Acids of a concentra- 
tion of 39.9, 40.8, and 41.4 per cent will dissolve 7, 12 to 13, and 
15 per cent of cellulose, respectively. If water is added to the 
solutions during the first 30 to 45 minutes, the cellulose is quanti- 
tatively precipitated. 

Hydrobromic acid of a concentration of 48 per cent gelatinizes 
cellulose. The latter is dissolved completely by 57 per cent acid 
at room temperature, and by 66 per cent acid at 0°. 

Concentrated hydriodic acid does not dissolve cellulose in the 
cold. 

Hydrofluoric acid of a concentration of 70 to 75 per cent gela- 
tinizes and dissolves cellulose rapidly. 2 

The reputed solvent power of chloral hydrate 3 is probably due 
to hydrochloric acid formed by decomposition. In the presence 
of chloral hydrate and concentrated sulphuric acid, cellulose gives 
dichloralglucoses. 4 

Sulphuric acid of a concentration of 70 to 75 per cent is most 
suitable for dissolving cellulose. The 80 per cent acid soon 
produces charring. While 60 per cent acid gelatinizes cellulose, 
it will not dissolve it at room temperature during 3 weeks. 5 

Numerous mixtures and concentrations of acids have been 
employed for dissolving cellulose for the production of filaments 
as well as for saccharification. There may be mentioned: 
concentrated sulphuric acid; 6 a mixture of sulphuric and phos- 
phoric acids; 7 a mixture of hydrochloric and sulphuric or phos- 
phoric acids; 8 and glacial acetic and phosphoric acids. 9 

1 Ferguson's table; only 38.35 per cent by that of Lunge and 
Marchlewski. 

2 L. Zechmeister, I.e., 38. 

3 T. A. Edison, Chem. News, 36 (1877), 138. 

4 J. H. Ross and J. M. Payne, J. Am. Chem. Soc, 45 (1923), 2363. 

5 H. Muhlmeister, Diss. Hannover. (1913), p. 21. 

6 R. Langhans, G. P. 72572 (1891). 

7 R. Langhans, G. P. 82857 (1893). 

8 Z. Ostenberg, U. S. P. 1218954 and 1242030 (1917); 1315393 (1919); 
1348731 (1920); International Cellulose Co. F. P. 484442 (1917); V. Hotten- 
roth, Danish P. 23957 (1918); Swiss P. 76329 (1917). 

9 K. H. Hofmann, G. P. 227198 (1909). 



254 CHEMISTRY OF CELLULOSE AND WOOD 

Berl 1 dissolves cellulose in 60 to 77 per cent sulphuric acid at a 
temperature not above —10° to obtain a solution suitable for 
films and artificial silk. 

Cellulose dissolves instantly in a solution of zinc chloride in 
twice its weight of concentrated hydrochloric acid. 2 A solution 
of zinc chloride in two parts of acetic anhydride gelatinizes cellu- 
lose but does not dissolve it even on heating. 3 In this case, 
acetylcellulose is formed. 

Arsenic acid slowly dissolves cellulose in the cold, solution 
being hastened by heating. 4 Selenic acid shows a slight solvent 
action on heating. 

Viscose. — In 1893, Cross and Bevan 5 published their discovery 
of the viscose reaction. They found that when cellulose was 
impregnated with a mercerizing solution of caustic soda, and 
carbon bisulphide then added, a water-soluble sodium salt of 
cellulose xanthogenic acid or thiocarbonic acid was obtained. 
This salt, called viscose, could be readily decomposed with the 
formation of a gelatinized cellulose. The regenerated cellulose 
was known as viscoid, though today the term is generally sup- 
planted by viscose. 

Cellulose can be gelatinized by the thiocarbonate reaction, 
and yet left insoluble in water, by using caustic soda of a strength 
insufficient to produce mercerization. 6 Cotton is treated with 
2.5 to 3 parts of 9 per cent sodium hydroxide and 15 per cent of 
carbon bisuphide. On the addition of water, the fibers swell in 
proportion to the ratio of alkali to cellulose employed, and the 
temperature at which the alkali reacts with the cellulose. De 
Cew 7 adds carbon bisulphide to cellulose that has absorbed 
approximately 2 per cent by weight of alkali. 

Theoretically, during the viscose reaction, sodium cellulose 
xanthogenate is formed as follows: 



1 E. Berl, G. P. 259248 (1912). 

2 C. F. Cross and E. J. Bevan, Chem. News, 63 (1891), 66. 

3 T. F. Hanausek, Chem. Ztg., 18 (1894), 441. 

4 H. G. Deming, J. Am. Chem. Soc, 33 (1911), 1523. 

5 C. F. Cross, E. J. Bevan, and C. Beadle, Ber., 26 (1893), 1090; E. P. 
8700 (1892); U. S. P. 520770 (1894). 

6 C. F. Cross and E. J. Bevan, E. P. 126174 (1918). 

7 J. A. de Cew, U. S. P. 1140799 (1915). 



GELATINIZED CELLULOSE 255 

(C 6 H 10 O 5 )x + zNaOH = (C 6 H 9 4 .ONa) x + zH 2 0. 
C 6 H 9 4 .ONa + CS 2 = C 6 H 9 4 .O.CS.SNa. 

Sodium cellulose 
xanthogenate 

In practice, it was found necessary to use twice the theoretical 
amount of sodium hydroxide in order to obtain a soluble product. 
This phenomenon, coupled with the inability of the authors 1 
to obtain higher than a dibenzoyl derivative of cellulose by the 
Baumann reaction, led them to suppose that two hydroxyl 
groups in the cellulose reacted with or were influenced by the 
alkali. 

Cellulose xanthogenic acid is a comparatively strong acid and 
its salts are not decomposed by weak organic acids. By titrating 
first with acetic acid, then with a mineral acid, only one sodium 
atom can be accounted for in each C 6 group. The same result 
is obtained by titrating, with iodine, pure sodium cellulose 
xanthogenate isolated by neutral dehydrating agents. 

2RO.CS.SNa + 21 = RO.CS.S-S.CS.OR + 2NaI 

The viscose solution decomposes spontaneously on standing, with 
the setting free of gelatinized cellulose. During this transition, 
three fairly well-defined stages can be observed. The maximum 
degree of reaction gives the C 6 xanthogenate, C 6 H 9 4 .O.CS.SNa. 
The cellulose residue, in reality, lies between C 6 and Ci 2 , since 
no sharp line of demarcation exists. The compound is not pre- 
cipitated from the freshly prepared solutions by dehydrating 
agents. After about 24 hours, it reverts to the Ci 2 xanthogenate 
that can be precipitated by such reagents as alcohol and salt, 
and redissolved in water. In the absence of an excess of alkali, 
the viscose solution changes in 6 to 10 days to the C 24 xantho- 
genate. It is soluble in alkaline solution but insoluble in water, 
since it is completely precipitated by acetic acid; hence the 
solubility is due to a reaction between the alkali and the hydroxyl 
groups of the cellulose. 2 

1 C. F. Cross, E. J. Bevan, and J. F. Briggs, Ber., 34 (1901), 1513. 

2 C. F. Cross and E. J. Bevan, " Cellulose" (1916), p. 318. 



256 CHEMISTRY OF CELLULOSE AND WOOD 

Ost 1 found that in freshly prepared viscose, two atoms of 
sulphur and two atoms of sodium react with the cellulose to 
form the compound C 6 H 8 3 (ONa)O.CS.SNa, thus confirming 
Cross and Be van. In "ripening" there is at first a progressive 
decrease in viscosity, owing to the depolymerizing action of the 
excess of alkali, then an increase in viscosity as coagulation sets 
in. The stability is increased by the presence of a certain amount 
of free caustic soda; 100 cubic centimeters of a viscose solution 
containing 2 grams of cellulose and 6 to 9 grams of free caustic 
soda remained unchanged for 3 months. A larger amount of 
alkali decreased the stability. A sample of viscose after standing 
5 days, and examined just prior to coagulation, had the composi- 
tion C 24 H 3 80 18 (ONa)O.CS.SNa. 

According to Vieweg, 2 the mercerization of cellulose causes 
several cellulose units to combine. Immersion of cellulose in 
18 per cent sodium hydroxide gives the compound (C 6 Hi O 5 )2.- 
NaOH. If the cellulose is pressed until it retains twice its 
weight of the alkaline solution, the compound (C 6 Hi O 5 )3.NaOH 
remains. The latter will combine with only 16 per cent of carbon 
bisulphide, based on the cellulose, so that the formula must be 
(C 6 Hio0 5 )3.CS2.NaOH. Titration with iodine showed that only 
the complexes (C 6 Hi O 5 )3 and (C 6 H 1005)4, especially the latter, 
exist in solutions capable of spinning. 

If ripening yields the C 2 4:CS 2 aggregate, then acid hydrolysis 
should yield more hydrogen suphide and less carbon bisulphide 
as ripening proceeds. Leuchs 3 found that this was not the case. 
Viscose ripened at periods varying from 2 to 6 days showed the 
ratio CS 2 : C 6 H w 5 to remain constant at 1 : 2. Hence the ripening 
process is considered to be colloidal rather than chemical. His 
methods are open to question. 

Subjection of cellulose acetate to the viscose reaction gives a 
xanthate of the composition C 6 H 5 (ONa)3.(O.CS.SNa)2. In 
contrast with cellulose xanthate, it accordingly appears to contain 
five active hydroxyl groups. 4 

1 H. Ost, F. Westhoff, and L. Gessner, Ann., 382 (1911), 349; cf. 
G. Bonwitt, Z. angew. chem., 26 (1913), 89. 

2 W. Vieweg, Zellstoff u. Papier, 2 (1922), 18. 

3 K. Leuchs, Chem. Zig., 47 (1923), 801. 

4 R. Wolffenstetn and E. Oeser, Ber., 56 (1923), 785. 



GELATINIZED CELLULOSE 257 

The inventors of the viscose process had in view the use of the 
viscous solution for coating fibers and fabrics. It remained for 
Stearn 1 to recognize that the progressive changes taking place 
in the ripening of viscose resulted in a product suitable for 
spinning filaments 2 and making films. 

Considerable latitude is shown in the commercial methods 
for preparing viscose. 3 According to the methods of prepara- 
tion, solutions having a wide range of properties can be obtained. 

Laboratory Preparation of Viscose. — Immerse 37 grams of sur- 
gical cotton in 558 cubic centimeters of 17.5 per cent sodium 
hydroxide (sp. gr. 1.2) at 20° for 1.5 hours. in a tightly stoppered 
bottle. Drain and press the cotton to a weight of 160 grams, 
pick it apart with forks, and preserve in a tightly stoppered 
bottle at 18° for 96 hours. To the alkali-cellulose in a closed 
rotating vessel, add 30 cubic centimeters of carbon bisulphide 
over a period of 2 hours. The color should be a deep orange 
yellow. Dissolve the cellulose xanthate by the gradual addition 
of 120 cubic centimeters of 4 per cent sodium hydroxide and 25 
cubic centimeters of 10 per cent sodium suphite, using mechanical 
stirring. Add further amounts of 4 per cent sodium hydroxide 
if necessary until the viscosity by the falling-sphere method shows 
a reading of 20 millimeters in 30 seconds. Filter the solution 
through light filter cloth at a pressure of 40 pounds to the square 
inch; then subject it to a vacuum of 20 millimeters for 4 hours 
to remove air bubbles. Store the solution so obtained in a 
tightly stoppered bottle at 18° for 90 to 150 hours to age. To 
prepare a film, the solution is spread uniformly on a glass plate 
and exposed for 2 minutes to a coagulant consisting of: glucose, 
10 per cent; sulphuric acid, 10 per cent; zinc sulphate, 1 per 
cent; anhydrous sodium sulphate, 14 per cent; and water, 65 
per cent. The sheet is then washed, bleached with potassium 

1 C. Stearn, J. Soc. Dyers Colourists, 19 (1903), 230; E. P. 1020, 1021, 
1022 (1898); 7023 (1903); U. S. P. 716778 (1902). 

2 The world's production of artificial silk, now called rayon, reached 
50,000 tons in 1923. Most of it is produced by the viscose process. In 
comparison with other artificial silks, viscose has the great advantage that 
the various manufacturing steps can be carried out entirely in aqueous 
solutions of cheap chemicals. 

3 F. J. G. Beltzer, Kunsistoffe, 2 (1912), 45; F. Becker, "Die Kuntseide" 
(1912), p. 270; B. M. Margosches, "Die Viskose" (1906). 



258 CHEMISTRY OF CELLULOSE AND WOOD 

permanganate followed by sulphurous acid, washed, and dried 
under tension. 1 

Sulphite pulp is used almost entirely in the commercial prep- 
aration of viscose. Bleached pulp is usually employed, since 
to produce a satisfactory spinning solution and final product, the 
pulp should be free from lignin and coloring matter. 2 During 
conversion, "pure" bleached wood pulp will lose as much as 20 
per cent of its weight, the largest loss being in the mercerization. 3 
Purified cotton cellulose, on the other hand, shows a slight 
increase in weight on regeneration. 4 

According to Beltzer, 5 the wood pulp is purified by heating in a 
tank of pitch pine with hydrofluoric acid of 0.5 to 1.0° Be. for 2 
to 3 hours at 80°. This removes the silica and other inorganic 
compounds, and renders the organic impurities more easily 
soluble in alkali. After draining and washing, the pulp is boiled 
in sodium hydroxide of 1 to 2° Be. for 3 to 4 hours, and again 
washed. 

Previous to the preparation of the alkali-cellulose, the cellulose 
should be in a state of fine subdivision. It is also essential to 
mill or knead the mixture of cellulose and alkali thoroughly to 
obtain a uniform reaction. 

A viscose solution of the highest viscosity is obtained by adding 
carbon bisulphide to the cellulose immediately after or during 
mercerization, if the temperature is kept low. According to the 
treatment given the alkali-cellulose or the viscose solution, 
solutions can be prepared having 5 and 20 per cent of cellulose 
and yet having the same viscosity. 6 Reduction in viscosity is 
obtained by storing the alkali-cellulose for a prolonged period 
or by heating it previous to the addition of the carbon bisulphide. 
The same end is also attained by a slight oxidation of the alkali- 
cellulose with sodium peroxide, or air in the presence of catalysts. 7 
As previously mentioned, reduction in viscosity is produced 

1 F. D. Snell, J. Ind. Eng. Chem., 17 (1925), 198; cf. E. H. Morse, Ibid., 
18 (1926), 398-400. 

2 L. P. Wilson, J. Soc. Chem. Ind., 39 (1920), 266R. 

3 C. Beadle, J. Franklin Inst., 143 (1897), 3. 

4 C. F. Cross and E. J. Bevan, "Cellulose" (1916), p. 28. 

5 F. J. G. Beltzer, Kunststoffe, 2 (1912), 42. 

6 C. Beadle, J. Franklin Inst., 138 (1894), 103. 

*L. P. Wilson, J. Soc. Chem. Ind., 39 (1920), 177T. 



GELATINIZED CELLULOSE 259 

by keeping the viscose solution at room temperature for several 
days. Ripening is hastened by heating, this result being secured 
by heating at 50 to 60° for 2 hours or for a shorter time at tem- 
peratures up to 90°. 

The use of hydrocellulose permits obtaining a viscose solution 
of lower viscosity and with a smaller consumption of chemicals. 
Wood pulp that has been heated with dilute mineral acids can 
be converted into viscose with only 30 to 40 per cent of the quanti- 
ties of sodium hydroxide and carbon bisulphide usually 
employed. 1 

Viscose solutions on standing coagulate spontaneously, 
forming a solid mass in the containing vessel. Shrinkage then 
sets in, a tough mass of gelatinized cellulose surrounded by a 
yellow alkaline solution being finally obtained. The coagulated 
viscose, if dilute, contracts to 10 per cent of cellulose and 90 
per cent of water. If stronger than 10 per cent it will take up 
water and expand. 2 The coagulum is not dehydrated by 
treatment with alcohol. 

The stability of viscose solutions is greatly increased by low 
temperatures. They are usually stored at temperatures 1 to 2° 
above zero. Coagulation requires about 14 days at 10° and under, 
and 6 to 10 days at 10 to 20°. 3 According to Beadle, 4 a solution 
that would keep 40 days at 1.7°, would coagulate in 25 days at 8.3°, 
or in 6 days at 18.3°. A thin film on glass will coagulate in a few 
minutes at a temperature of 121.° Small amounts of viscose 
coagulate more rapidly than large amounts at a given tempera- 
ture. Coagulation is also influenced by the amount of free alkali 
present in the solution. 

Addition of strong mineral acids decomposes the viscose and 
causes immediate coagulation. Acetic acid, on the other hand, 
will neutralize the free alkali, but will not decompose the viscose. 
Alcohol and solutions of salts, such as sodium chloride, precipi- 
tate sodium cellulose-xanthogenate that can be redissolved in 
water. These properties can be used in the purification of 
viscose. For example, the free alkali can be neutralized with 

1 W. Cross, E. P. 185433 (1921); Cj. C. F. Cross, G. P. 92590 (1896). 

2 C. Beadle, I.e., 105. 

3 C. F. Cross, cited by B. M. Margosches, "Die Viskose," p. 70. 

4 C. Beadle, I.e., 104. 



260 CHEMISTRY OF CELLULOSE AND WOOD 

acetic acid and the viscose precipitated with alcohol. On 
washing the precipitate with alcohol most of the sodium acetate 
and other impurities are removed. Resolution in water is 
greatly aided by the presence of sodium hydroxide. 

The magnesium, sodium, and potassium salts of cellulose 
xanthogenate are soluble in water. The metal salts are insoluble 
and exhibit a variety of colors, the latter varying with the purity 
of the viscose solution. 

The suitability of a viscose for the preparation of artificial 
silk may be determined by forcing a fine stream of the solution into 
40 per cent acetic acid. If properly matured, coagulation takes 
place at once; otherwise the viscose passes into solution. Besides 
the sodium cellulose-xanthogenate, sulphides, polysulphides, 
and thiocarbonates are present as impurities. It has not been 
found expedient, however, to purify the solution before spinning. 1 
The viscose solution is forced by high pressure through fine 
orifices into a coagulating bath. Sulphuric acid and sodium 
acid sulphate are usually used as coagulants on account of their 
cheapness, in which case gelatinized cellulose is obtained at once. 
A modification consists in forming a thread of coagulated sodium 
cellulose-xanthogenate by means of salts, followed by a second 
bath to convert it to cellulose. 

The artificial silks, particularly viscose, have the disadvantage 
of greatly reduced strength in the wet state. To overcome this 
defect, the fiber is treated with formaldehyde in the presence of 
sodium hydroxide 2 or acids, 3 a condensation taking place between 
the aldehyde and the cellulose. This is known as the "stheno- 
sage " 4 or strengthening process. The sthenosized fiber is harder, 
and in the dry state shows a slight increase in tensile strength, but 
decreased elasticity. In the wet state, in comparison with the 
untreated fiber, the tensile strength is two to three times greater. 
The difficulty in stopping the reaction at the point of optimum 
strength and elasticity has apparently prevented the operation 
of the process on an industrial scale. 5 

1 V. Hottenroth, " Ullmann Enzykl. Techn. Chem./' 7 (1919), 335. 

2 E. R. Blumer G. P. 179590 (1906). 

3 X. Eschalier, F. P. 374724 (1906). 

4 TZdepos, strength. 

5 V. Hottenroth, I.e., 323. 



GELATINIZED CELLULOSE 261 

Sthenosized fibers cannot be mercerized or made to undergo 
the viscose reaction. 1 They are not stained blue by iodine- 
potassium iodide solution and are only feebly colored by direct 

I 
HC.(X 

cotton dyes. A true formal derivative, /CH 2 , appears 

HC.CK 

J 

to be formed, since the whole of the combined formaldehyde 
can be recovered by heating with dilute sulphuric acid. Accord- 
ing to Samec, 2 the formaldehyde first reacts with a hydroxyl 

I 
group to form the group — C.OCH 2 OH that, in turn, reacts 

I 
with an adjacent hydroxyl group to form an inner anhydride. 
The fixation of formaldehyde appears to increase with the 
degree of gelatinization, though large quantities are taken up by 
hydrocellulose and oxycellulose. 

1 F. J. G. Beltzer, Eighth Int. Cong. Appl. Chem., N. Y., 7 (1912), 9. 

2 M. Samec and S. Ferjancic, Kolloidchem. Beihefte, 14 (1921), 209. 



CHAPTER VIII 
OXYCELLULOSE 

The recognition of oxycellulose as a distinct modification of 
cellulose is due to Witz. 1 While examining cotton fabrics that 
had been accidentally overbleached, he noticed that the edges 
of the resulting holes had a marked affinity for basic dyes, 
particularly methylene blue. The characteristic product could 
be obtained by the use of oxidizing agents, such as chlorine, 
potassium permanganate, hydrogen peroxide, and chloric and 
chromic acids. Analysis showed it to contain less carbon and 
more oxygen than normal cellulose — hence it was given the name 
oxycellulose. 2 

Formation with Nitric Acid. — Cellulose when boiled with 60 
per cent nitric acid gave, in addition to oxalic acid, about 30 
per cent of oxycellulose. 3 On removing the acid with hot water, 
the mass became gelatinous and resembled horn after drying. 
It was entirely soluble in dilute alkali and could be precipitated 
unchanged by acids, alcohol, salt solutions, and even by strong 
solutions of the caustic alkalis. As a result of a difference in 
properties, Cross and Bevan 4 called their product, obtained with 
nitric acid, /3-oxycellulose, and that of Witz with bleaching 
powder, a-oxy cellulose. 

The gelatinous product obtained by Sacc 5 by boiling spruce 
wood with nitric acid was believed to be pectic acid. Lindsey 
and Tollens, 6 who repeated Sacc's work, obtained a gelatinous 
oxycellulose completely soluble in dilute ammonia and in very 
dilute caustic soda; strong caustic soda precipitated it as a jelly. 

1 G. Witz, Bull Rouen, 10 (1882), 416-464; 11 (1883), 169-232. 

2 G. Witz, I.e., 448. 

3 C. F. Cross and E. J. Bevan, J. Chem. Soc., 43 (1883), 22-23. 

4 C. F. Cross and E. J. Bevan, J. Soc. Chem. Ind., 3 (1884), 206. 

5 F. Sacc, Ann. chim. phys., [3] 25 (1848), 218. 

6 J. B. Lindsey and B. Tollens, Ann., 267 (1892), 366-370. 

262 



OX YCELL ULOSE 263 

Using the procedure of Cross and Bevan, von Faber and 
Tollens 1 obtained a yield of 70 per cent of oxy cellulose from 
cotton. Nastjukoff 2 obtained a yield of 90 per cent of /3-oxy- 
cellulose, completely soluble in hot ammonium hydroxide, by 
heating 1 part of cellulose with 2.5 parts of nitric acid (sp. gr. 
1.3) on the water bath for 1 hour. Bull 3 prepared this product by 
heating 75 grams of cotton and 750 grams of 60 per cent nitric 
acid in a stoppered retort on a water bath for 24 hours and 
returning the distillate at intervals. After filtering, washing with 
nitric acid, 60 per cent alcohol, 90 per cent alcohol, and finally 
with ether, the yield was 35 per cent. The oxycellulose dissolved 
completely in water and gave opalescent solutions with aqueous 
solutions of ammonia, pyridine, piperidine, sodium carbonate, 
and weak solutions of sodium and potassium hydroxides. Excess 
of sodium hydroxide produced gelatinization, while a large 
excess produced a clear and limpid mass. Acids precipitated 
it in a gelatinous state. The substance which was regenerated 
from the nitrate of the oxycellulose with ammonium sulphide 
was apparently not /3-oxycellulose, since it no longer swelled in 
water and dissolved to only a slight extent in dilute sodium 
hydroxide. This behavior is difficult to explain unless the original 
oxycellulose consisted largely of a material related to Guignet's 
soluble cellulose 4 and was rendered insoluble by the strong 
nitrating acids. 

The oxycellulose when freshly prepared with nitric acid is not 
colored by iodine or by Schultze's reagent (zinc chloride and 
iodine-potassium iodide), but the horny mass to which it dries 
is colored deep blue by the latter. 5 According to others, the fresh 
material is also colored bluish violet. 6 Iodine and sulphuric acid 
color oxycellulose (by chloric acid) more rapidly and distinctly 
than it does cellulose. 7 

Oxycellulose by Chloric Acid. — When paper has been written 
upon with chloric acid or aluminum chlorate, dried at 50 to 60°, 

1 O. von Faber and B. Tollens, Ber,, 32 (1899), 2594. 

2 A. Nastjukoff, Ber., 34 (1901), 3589. 

3 B. S. Bull, J. Chem. Soc., 71 (1897), 1090-1097. 

4 C. E. Guignet, Compt. rend., 108 (1889), 1258. 

5 C. F. Cross and E. J. Bevan, J. Chem. Soc., 43 (1883), 23. 

6 B. S. Bull, I.e.; J. B. Lindsey and B. Tollens, I.e., 368. 

7 L. Vignon, Compt. rend., 125 (1897), 449. 



264 CHEMISTRY OF CELLULOSE AND WOOD 

washed, and then dyed with methylene blue, the writing is 
colored a dark purplish blue. 1 

Vignon 2 considered the best method of preparing oxycellulose 
to be as follows: 30 grams of purified cotton are placed in a 
boiling solution of 150 grams of potassium chlorate in 3 liters 
of water, and 125 cubic centimeters of 35 per cent hydrochloric 
acid are gradually added. After maintaining at the boiling 
point for an hour, the oxycellulose is washed with water by 
decantation until free from acid, then with alcohol, and air 
dried. The yield is 96 per cent. 3 

On treating oxycellulose prepared as above with a 28 per cent 
solution of potassium hydroxide and filtering, there was obtained 
a residue, amounting to 60 per cent, having the properties of 
cellulose. 4 Slow addition of the filtrate to a cooled solution of 
hydrochloric acid caused separation of 10 per cent of a white 
substance called "soluble cellulose." Except for its solubility 
in alkali, it is not apparent why the product should have been 
so named, since its solubility in 1 liter of cold and of boiling 
water was only 0.02 and 0.4 gram, respectively. Ultimate analy- 
sis showed it to have the composition of cellulose, but it differed 
therefrom in its heat of formation and other properties. It 
contained 3.5 per cent moisture, gave about 5.0 per cent furfural 
on distillation with hydrochloric acid, reduced Fehling's solu- 
tion, and gave a rose color with Setoff's reagent. It was soluble 
in the caustic alkalis, partially in hydrochloric acid, and com- 
pletely so in nitric acid. The product probably consists of a 
mixture of oxycellulose, hydrocellulose, and colloidal cellulose. 

Oxycellulose by Chromic Acid. — The oxycellulose is prepared 
as follows: 40 grams of cotton are placed in a solution of 60 
grams of potassium bichromate and 80 grams of sulphuric acid 
in 3 liters of water, and boiled for 1 hour. There is obtained 69 
per cent of a green powder with 2.24 per cent ash. 5 A similar 

!H. Schmid, Wagner's Jahresber., 29 (1883), 1076-1078; cf. G. Witz, 
Bull. Rouen, 11 (1883), 201. 

2 L. Vignon, Compt. rend., 125 (1897), 449. 

3 E. Heuser and F. Stockigt, Cellulosechemie, 3 (1922), 66. 

4 L. Vignon, Compt. rend., 136 (1903), 969-970. 

5 L. Vignon, Bull. soc. chim., 19 (1898), 811; cf. G. Witz, Bull. Rouen, 
11 (1883), 207. 



OXYCELLULOSE 265 

high retention of chromium oxide was observed by Cross and 
Bevan. 1 Heuser 2 obtained a yield of 83 per cent of oxycellulose 
having 0.56 per cent ash after washing with hot 2 per cent sul- 
phuric acid. 

Oxycellulose by Permanganates. — The oxidizing action on 
cellulose of potassium permanganate alone, and with the addition 
of acids and alkalis, has been studied by various investigators. 3 
As pointed out by Witz, oxidation in alkaline solution tends to 
the formation of completely soluble products rather than 
oxycellulose. 4 

An oxycellulose having high reducing properties is prepared 
as follows: 5 30 grams of filter paper are treated with 600 to 900 
cubic centimeters of 6.5 per cent sodium hydroxide solution and 
shaken until a pulp is formed. The pulp is then thinned with 
water, rendered acid with sulphuric acid, filtered off by suction, 
and washed with water. The moist pulp is suspended in 600 
to 900 cubic centimeters of 7.5 per cent sulphuric acid and 
rapidly stirred mechanically. By means of a separatory funnel, 
12 grams of potassium permanganate that has been dissolved in 
7.5 per cent sulphuric acid are slowly added. If 1 to 2 hours are 
taken for the addition, the solution is completely decolorized in 
4 hours from the start. The finely divided brown precipitate is 
filtered off, suspended in dilute suphuric acid, and decolorized 
with hydrogen peroxide. It is then washed with water until free 
from acid and manganese sulphate, and dried at a temperature 
not exceeding 40° to prevent the formation of a horny mass. 
The yield is 94.5 per cent. 

The oxycellulose on heating with sodium hydroxide solution 
(8 to 10 per cent) on the water bath for 5 hours gives a golden- 
yellow solution changing to brown. Probably as a result of 
oxidation, only a slight precipitate is obtained on acidification. 
It dissolves to a deep-yellow solution in 5N sodium hydroxide; 

1 C. F. Cross and E. J. Bevan, Chem. Neios, 64 (1891), 63. 

2 E. Heuser and F. Stockigt, Cellulosechemie, 3 (1922), 66. 

3 G. Witz, Bull. Rouen, 11 (1883), 211; A. Nastjukoff, Ber., 33 (1900), 
2237; 34 (1901), 720; J. Allan, J. Soc. Dyers Colourists, 14 (1898), 6; H. 
Moore, Ibid., 31 (1915), 180. 

4 J. Allan, J. Soc. Dyers Colourists, 14 (1898), 6. 

5 E. Knecht and L. Thompson, J. Soc. Dyers Colourists, 36 (1920), 251; 
cf. E. Heuser and F. Stockigt, Cellulosechemie, 3 (1922), 66. 



266 CHEMISTRY OF CELLULOSE AND WOOD 

this on neutralization gives an acid cellulose of only slight reduc- 
ing power. 

Knecht's oxycellulose has a copper number of 14.2 compared 
with 7.6 for Nastjukoff's oxycellulose prepared with bleaching 
powder. It has strong aldehydic properties but is only slightly 
acidic. It reduces solutions of silver, cupric and ferric salts, 
and, especially in the presence of alkalis, such dyes as indigo, 
methylene blue, safranine, and rosinduline. 

On following, by means of the copper numbers, the action on 
cotton of gradually increased amounts of potassium permanga- 
nate in 7.5 per cent sulphuric acid, it was found that initially the 
reducing value was proportional to the oxygen consumed. With 
one-half atom of oxygen the copper number was 12.7, while with 
one atom of oxygen this value increased to only 14.0, indicating 
that while aldehydic and ketonic groups are formed for the most 
part, further oxidation then takes place to the carbonyl group. 1 

Oxycellulose by Hypochlorites. — NastjukofP used solutions of 
bleaching powder of 4 and 12° Be. With the stronger solution 
the oxidized cellulose was almost completely soluble in a 10 per 
cent solution of sodium hydroxide. 

The oxycellulose prepared by Qvist 3 from sulphite pulp with 
sodium hypochlorite had the following properties: 

Copper number (uncorrected) 34 . 6 

Solubility in 10 per cent NaOH 81.7 per cent 

Moisture 9 . 52 per cent 

In the case of hypochlorites, moderate additions of acid decrease 
the amount of oxycellulose formed, while with alkali a minimum 
amount of oxycellulose is formed at a concentration of 1 gram 
of sodium hydroxide to a liter of sodium hypochlorite containing 
11.47 grams of chlorine. 4 The conversion of cellulose to oxy- 
cellulose by hypochlorites is promoted greatly by the presence 
of oxides of mercury, copper, cobalt, nickel, iron, and chromium. 5 

1 E. Knecht, J. Soc. Dyers Colourists, 38 (1922), 132-136. 

2 A. Nastjukoff, Ber., 33 (1900), 2237. 

3 W. Qvist, Papier-Fabr., 17 (1919), 821. 

4 H. Moore, J. Soc. Dyers Colourists, 31 (1915), 180. 

5 P. Weyrtch, Z. angew. Chem., 28 (1915), Ref. 399; W. Thomson, 
J. Soc. Dyers Colourists, 30 (1914), 142. 



OXYCELLULOSE 267 

Oxycellulose has also been prepared by the action of bromine 
in the presence of calcium carbonate. 1 Dry bromine is without 
action on dry cellulose. ? This is true also of the other halogens. 
According to Klemm, 3 iodine is without chemical action on 
cellulose in either a wet or a dry condition. Cotton and wood 
pulps absorb appreciable quantities of iodine; if, after standing 
some hours, sodium hydroxide is added and the mixture steam 
distilled, appreciable quantities of iodoform are obtained. 4 
The residual cellulose on re treatment continues to form iodoform. 
Wood cellulose gives more iodoform than cotton; 10 grams of 
poplar pulp gave 0.15 gram of iodoform. Collie 5 obtained carbon 
tetrabromide by treating cellulose with a hypobromite and 
distilling. 

Oxycellulose by Salts and Air. — It has been stated that oxy- 
cellulose is produced by heating moist cotton at 50° with salts 
of copper, cobalt, manganese, nickel, iron, and chromium, their 
activity being in the order mentioned. 6 Certain salts also form 
hydrocellulose under the same conditions. 

Paper coated with ink rich in iron oxide, or when coated with 
colloidal ferric oxide, is converted into a powder having a micro- 
crystalline structure by heating in a current of dry air at 50 to 
60° for 5 to 6 hours. 7 According to Bolton and Doree, 8 no tender- 
ing takes place with colloidal ferric oxide prepared from ferric 
chloride when it is free from hydrochloric acid. In the presence 
of the acid, hydrocellulose is formed. It is a common observa- 
tion, however, that when a cotton string is attached to iron and 
left in a moist atmosphere, the string in time becomes impreg- 
nated with rust and loses strength at the point of attachment. 

Calico mordanted with chromium, and then treated with warm 
hydrogen peroxide in the presence of ammonia or sodium 



1 O. von Faber and B. Tollens, Ber., 32 (1899), 2589. 

2 A. P. N. Franchimont, Rec. trav. chim., 2 (1883), 91. 

3 P. Klemm, Wochbl. Papierfabr., 43 (1912), 464. 

4 J. Huebner and J. N. Sinha, J. Soc. Chem. Ind., 42 (1923), 255-260T. 

5 J. N. Collie, J. Chem. Soc, 65 (1894), 262-264. 

6 L. L. LIoyd, J. Soc. Dyers Colourists, 26 (1910), 273. 

7 K. Haertling, Kolloid Z., 25 (1919), 74-79. 

8 H. S. Bolton and C. Doree, J. Soc. Dyers Colourists, 40 (1924), 292. 



268 



CHEMISTRY OF CELLULOSE AND WOOD 



hydroxide, gradually disintegrates, with the formation of oxycel- 
lulose. In the absence of chromium, there is no material change. 1 

A salt such as mercuric chloride is reduced by cellulose, appar- 
ently with the formation of oxy cellulose. 2 

Oxycellulose by Alkalis and Air. — Blondel 3 observed that the 
hydroxides and carbonates of the alkalis in the presence of air 
caused a weakening of cotton and produced a greater affinity 
for dyes, with the exception of the azo colors. 

Cotton in the presence of alkali and air may be severely 
oxidized. The following table shows the losses suffered by 
cotton that had been scoured, bleached, and dried at 85°: 





Loss in per cent with 


5 per cent of 


Oxygen 


Steam and 
oxygen 


Air 


Steam and 
air 


NaOH 

KOH 


11.05 

22.8 

8.2 

13.7 

5.9 

3.1 


17.3 
29 8 
10.1 
16.4 
6.8 
3.5 


5.2 
8.4 
3.9 
5.3 
2.2 
2.05 


9.2 
11.7 


Na 2 C0 3 


5 4 


K 2 C0 3 


6.9 


NaoB 4 7 


2.8 


Na 2 HP0 4 


2.3 







The residual fiber was weak, owing to the presence of oxycellu- 
lose. 4 Thies 5 found a very decided formation of oxycellulose 
when cotton was treated with concentrated alkalis saturated 
with oxygen. The fiber decreased about 80 per cent in strength. 
Oxycellulose by Hydrogen Peroxide. — The formation of 
oxycellulose with hydrogen peroxide was first studied by Witz. 6 
The action on cotton is greatly increased in the presence of the 
oxides of iron, chromium, aluminum, and magnesium, those of 



1 E. Knecht, J. Soc. Dyers Colourists, 13 (1897), 109. 
2 L. Vignon, Compt. rend., 116 (1893), 517, 645. 

3 E. Blondel, Bull. Rouen, 10 (1882), 472. 

4 C. O. Weber, J. Soc. Chem. Ind., 12 (1893), 118-119; cf. S. R. Trotman 
and S. J. Pentecost, Ibid., 29 (1910), 4. 

5 F. H. Thies, Farber-Ztg., 24 (1913), 393. 

6 G. Witz, Bull. Rouen, 11 (1883), 197. 



OXYCELLULOSE 269 

iron and chromium being particularly effective. 1 The alteration 
of the cellulose is much greater if the cotton has been mercerized 
previous to treatment with hydrogen peroxide. The fiber 
disintegrates completely if sufficient sodium hydroxide is added 
to the oxidizing solution to bring it to 5 to 6° Be. 

By the action of two parts of 50 per cent hydrogen peroxide on 
cellulose for 2 months, Bumcke and Wolffenstein 2 obtained a 
product that they named "hydralcellulose." It reduced Fehl- 
ing's solution and ammoniacal silver solution, colored Schiff's 
reagent, and formed hydrazones containing 1.71 per cent of 
nitrogen, thereby showing its aldehydic properties. On treat- 
ment with alkali, the aldehyde presumably followed the Canni- 
zarro reaction, being converted into an insoluble alcohol, 
cellulose, and a soluble acid, acid cellulose, the latter having no 
aldehydic properties. There was formed one part of acid cellulose 
to two parts of cellulose. 

Acid cellulose dissolves in concentrated hydrochloric acid, from 
which it can be precipitated unchanged by water. On heating 
the solution or allowing it to stand, hydrolysis takes place very 
readily. Drying produces loss of solubility, especially in sodium 
hydroxide, and formation of a hard, brittle mass having the 
lactone formula, C 3 6H 6 o0 3 i. Acid cellulose may also be formed 
by boiling cellulose with a 30 per cent sodium hydroxide solution 
for an hour. 

Acid cellulose is sufficiently acid to decolorize phenolphthalein. 
Cellulose, hydralcellulose, and acid cellulose lactone from 
Schweizer's reagent and from boiling sodium hydroxide solution 
were nitrated by heating with four parts of nitric acid (86 per 
cent) at 85°. Practically identical products were obtained when 
examined as to: nitrogen content (8 per cent) of the portions 
soluble in alcohol and acetone; explosion point; and optical rota- 
tion. The molecular weight of the alcohol-soluble portion of the 
nitrated celluloses, determined in boiling acetone, gave an average 
molecular weight of 1350, so that hydralcellulose has the formula 
6C 6 H t0 O 5 + H 2 0. 

1 M. Prud'homme, Bull. Mulhouse, 61 (1891), 508; Compt. rend., 112 
(1891), 1374. 

2 G. Bumcke and R. Wolffenstein, Ber., 32 (1899), 2493-2507; cf. Ibid., 
34 (1901), 2415-2417. 



270 CHEMISTRY OF CELLULOSE AND WOOD 

The assumption of the hydrolysis of cellulose to aldehydic 
hydralcellulose by hydrogen peroxide is based on the work of 
Wurster, 1 who states that cane sugar is first inverted by hydrogen 
peroxide, then oxidized with evolution of carbon dioxide. Cel- 
lulose is scarcely attacked by hydrogen peroxide at low tempera- 
tures, but rapidly at the boiling point in alkaline or acid 
solution, with the formation of glucose and dextrin. According 
to Haller, 2 vegetable fibers are dissolved by the prolonged action 
of hydrogen peroxide, with the formation of dextrose, cellulose 
peroxide being an intermediate product. Chromium oxide 
hastens the decomposition, while aluminum oxide retards it. 

Haeussermann 3 claims that the acid cellulose of Bumcke and 
Wolffenstein does not contain a carboxyl group, while Bay 4 
believes it to be a mixture of hydrated cellulose and oxy cellulose. 
In view of our present knowledge of hydralcellulose it should be 
classed as an oxy cellulose. 

Oxycellulose prepared with hydrogen peroxide sometimes has 
an irritating action on the skin. 5 

Oxycellulose from Cuprammonium Solution. — Cuprammo- 
nium solution behaves towards cotton like hydrogen peroxide. 
Even though the solution is not sufficiently strong to dissolve 
the cellulose, the latter is oxidized and colored strongly by 
methylene blue. 6 According to Bumcke and Wolff enstein, 7 the 
cellulose precipitated from Schweizer's reagent is an acid cellulose. 

Oxycellulose by Reduction of Cellulose Nitrate. — According 
to Vignon, 8 the product of the reduction of the cellulose nitrates 
with ferrous chloride is oxycellulose and not cellulose. Reduc- 
tion with ammonium sulphide gives a product devoid of reducing 
properties. It is possible that in this case the reducing bodies 
are dissolved by the alkaline ammonium sulphide. Various 
products showed the following reduction with Pasteur's solution: 



1 C. Wurster, Ber., 22 (1889), Ref. 145. 

2 R. Haller, Chem. Zentr., 91, IV (1920), 543. 

3 C. Haeussermann, Z. Schiess-u. Sprengstoff, 1 (1906), 39, 305. 

4 C. Bay, Diss. Giessen (1913). 

5 C. G. Schwalbe and E. Becker, Zellstoff u. Papier, 1 (1921), 102. 
e M. Prud'homme, Compt. rend., 112 (1891), 1374. 

7 L.c. 

8 L. Vignon, Compt. rend., 131 (1900), 530. 



OXYCELLULOSE 271 

Grams CuO 

per 100 Grams 

Substance 

Oxycellulose 10 . 36 

Cellulose nitrate 13 . 71 

Oxycellulose reduced by FeGl 2 10 . 42 

Cellulose nitrate reduced by FeCl 2 11 . 70 

Oxycellulose nitrate reduced by FeCl 2 10.83 

Cellulose nitrate is a derivative of oxycellulose. 1 Dried at 80° 
it had the following composition : 

Per Cent 

C 23.89 

H 2.44 

N 13.94 

59.73 

These results agree with the formula: 

3[C 6 H 7 (N0 2 ) 3 5 ] + C 6 H 7 (N0 2 ) 3 6 = C 24 H 2S (N0 2 ) 12 21 . 

Berl and Fodor 2 likewise believe that cellulose nitrate is a deriva- 
tive of oxycellulose. Obviously, little reliance can be placed 
on ultimate analysis in detecting the presence of an extra atom 
of oxygen in so large a molecule. 

Oxycellulose by Electrolysis. — Cellulose in contact with the 
anode is converted into oxycellulose during the electrolysis of 
aqueous solutions. 3 

When a neutral 15 per cent solution of potassium chloride 
containing cellulose is electrolyzed, the cellulose is ultimately 
converted into soluble products. If the reaction is stopped after 
60 to 70 per cent of the cellulose has disappeared, the residue 
dissolves completely in 10 per cent sodium hydroxide. On 
prolonging the treatment, the oxycellulose forms a stable col- 
loidal solution in water. The oxycellulose showed a low viscosity 
by Ost's method 4 and copper numbers as high as 39.5. Hydrol- 
ysis with 72 per cent sulphuric acid gave 90 per cent of glucose. 
The high "hydrolysis difference" number, and other properties 

1 h. Vignon, Compt. rend., 131 (1900), 509; 136 (1903), 898. 

2 E. Berl and A. Fodor, Z. Schiess-u. Sprengstoff, 5 (1910), 254, 269. 

3 G. Gopplsroeder, Bull. Mulhouse, 45 (1875), 229; J. J. Hummel, 
J. Soc. Chem. Ind., 4 (1885), 306; G. Witz, Bull. Rouen, 11 (1883), 196. 

4 H. Ost, Z. angew. Chem., 24 (1911), 182. 



272 CHEMISTRY OF CELLULOSE AND WOOD 

indicated that the product was a hydro-oxy cellulose. 1 Cellulose 
is also converted into hydro-oxycellulose by treatment with 
chlorine and the hydrochloric acid resulting therefrom. 2 It is 
apparent that this product may be formed by any acid oxidizing 
agent. 

Cellulose Peroxide. — In 1906, attention was called by Cross 
and Bevan 3 to the fact that cellulose which had been bleached 
without the subsequent use of an antichlor apparently formed 
a cellulose peroxide. The active oxygen was retained for a 
long time. Iodine was liberated from potassium iodide solution 
and this property was not destroyed by washing with water or 
by neutralizing the acidity of the product. Boiling water or an 
antichlor destroyed its activity. The active oxygen disappeared 
on heating at 100°, leaving the fiber in the form of oxy cellulose. 
The peroxide reaction does not always disappear on drying, and 
apparently does not result in a decrease of the strength of linen 
yarn. 4 

The phenomenon is characteristic of bleached pulps, and is 
apparently not due to the presence of residual chlorine or hypo- 
chlorite, 5 as suggested by Zimmermann. 6 

A cellulose peroxide is also produced by the careful heating of 
cellulose in an acid solution of a persulphate, particularly 
ammonium persulphate, and allowing to cool. The compound 
decomposes at 100° with the formation of oxygen, carbon dioxide, 
and a very irritating gas. 7 The amount of active oxygen fixed 
is only 0.015 per cent. When suspended in water and methyl 
orange added, the ordinary yellow color results; but on the addi- 
tion of a neutral salt, such as sodium chloride, the liquid turns 
red. This shows the acid character of the product and the 
presence of oxy cellulose. 8 

1 R. Oertel, Z. angew. Chem., 26 (1913), 246-250; Chem. Ztg., 35 (1911), 
713. 

2 C. C. Budde, E. P. 184610 (1921). 

3 C. F. Cross and E. J. Bevan, Z. angew. Chem., 19 (1906), 2101-2102; 
Papier-Ztg., 32 (1907), 87-88. 

4 J. L. Heinke, Chem. Ztg., 31 (1907), 974. 

5 K. Rieth, Wochbl. Papierfabr., 38 (1907), 394: Montanus, Ibid., 1041. 

6 D. M. R. Zimmermann, Z. angew. Chem., 20 (1907), 1280. 

7 H. Ditz, Chem. Ztg., 31 (1907), 833-834, 857-885. 

8 H. Ditz, Chem. Ztg., 31 (1907), 844-845. 



OXY CELLULOSE 273 

Cotton treated with a solution of 20 grams of ammonium 
persulphate per liter, dried, and steamed for 134 hours decreased 
in strength 41 per cent. 1 

Kolb 2 observed that linen exposed to the action of ozone for 
a week was attacked; and Witz 3 , by means of methylene blue, 
detected the presence of oxy cellulose. Moist filter paper exposed 
to a stream of ozone of a concentration of 2 per cent showed no 
rise in temperature. 4 

Ozone in concentrations of 1 to 2 per cent rapidly attacks 
cotton with the formation of cellulose peroxide, an acid derivative, 
and some carbon dioxide. 5 The ozonized fiber rapidly reduces 
Fehling's solution but not ammoniacal silver solution. In 
the absence of water only a small amount of peroxide is formed, 
but much more in air-dried cotton. 6 The activity of cellulose 
peroxide soon disappears in air, but remains for some weeks in 
the product kept in a dry atmosphere. It is slowly decomposed 
by water, with the formation of hydrogen peroxide. It is 
decomposed to the extent of 25 per cent on heating for 2 hours 
at 37°, and almost completely after 2 hours at 95°. After 18 
hours' exposure to ozone, various forms of cellulose contained the 
following amounts of "active ozone": 

Per Cent 

Cotton (air dry) . 0056 

Mercerized cotton . 0106 

Lustra-cellulose . 0248 

Cellulose peroxide, through liberation of hydrogen peroxide, acts 
strongly on photographic plates, negative images of the fibers 
being obtained after 6 hours' exposure at 37°. 

Cotton and mercerized cotton, after continuous exposure for 
24 hours to ozonized oxygen containing 1.5 to 2.0 per cent ozone, 
had the following properties: 7 

1 A, Scheurer, Bull. Mulhouse, 71 (1901), 182. 

2 J. Kolb, Bull Mulhouse, 38 (1868), 914. 

3 G. Witz, Bull. Rouen, 11 (1883), 187. 

4 E. Erdmann and H. Stoltzenberg, Chem. Zentr., 79 ,11 (1908), 457. 

5 M. Cunningham and C. Doree, J. Chem. Soc., 101 (1912), 497-512. 

6 C. Doree, J. Chem. Soc., 103 (1913), 1347. 

7 C. Doree, J. Soc. Dyers Colourists, 29 (1913), 208. 



274 



CHEMISTRY OF CELLULOSE AND WOOD 





Cotton 


Mercerized cotton 




Normal 


Ozonized 


Normal 


Ozonized 


Loss in weight, per cent 

Carbon, per cent 


44.4 

0.3 
1.2 


12.0 
43.5 

2.1 
16.9 

64.0 


43.2 

0.4 
1.7 


7.0 
43.5 


Methylene blue absorption, per 
cent 


3.1 


Copper number 


24.0 


Loss in 10 per cent KOH, per 
cent 


57.0 







Properties of Oxycellulose. — The property of cellulose peroxide 
of liberating iodine from a solution of potassium iodide is usually 
used for its detection. 

The heats of combustion of oxycellulose in comparison with 
other carbohydrates are given below. 1 The heat of combustion 
of the residue from the treatment of the oxycellulose with potas- 
sium hydroxide agrees well with that for cellulose. 

Calories 

Oxycellulose (by KC10 3 + HC1) 4124-4133 

Oxycellulose exhausted by KOH of 30° Be 4201 

Portion precipitated by acid 3929 

Cellulose 4190-4225 

Sucrose 3955 

Purified cotton (dried at 100°) 4223 

Mercerized cellulose* 3980 

Cellulose precipitated from Schweizer's reagent* 3982 

* Dried in vacuo over H2SO4. 

Oxycellulose has a marked affinity for basic dyes, 2 but only a 
very slight one for tetrazoic dyes. According to Harrison, 3 
the dyeing properties of oxycellulose are due mainly to the 
colloidal state of the cellulose portion. This accounts for the 
fact that certain forms of hydrocellulose and mercerized cellulose 
dye with methylene blue in a manner similar to oxycellulose. 

!L. Vignon, Compt. rend., 125 (1897), 450; 127 (1898), 873; 131 (1900), 
709. 

2 G. Witz, Bull. Rouen, 10 (1882), 455; L. Vignon, Compt. rend., 125 
(1897), 550; 131 (1900), 558. 

3 W. Harrison, J. Soc. Dyers Colourists, 28 (1912), 360. 



OX YCELL ULOSE 275 

Oxycellulose has the power of adsorbing vanadium salts from 
solutions as dilute as one part per million, as can be shown by 
the subsequent effect of the fiber in developing aniline black. 1 
It, however, shows no such intense adsorption for copper, alu- 
minum, and iron mordants. The extent of adsorption increases in 
the order given. 2 

There appears to be no industrial use for oxycellulose. It 
has been suggested for calico printing 3 and as a binding or coating 
agent 4 for paper. An attempt to use an alkaline solution as a 
mordant for basic dyes gave negative results. 5 In paper making, 
the presence of a certain amount of oxycellulose facilitates gelat- 
inization of the fibers by beating. 

Effect of Heat. — The statement 6 that hydrocellulose becomes 
brown at 130 to 150°, at which temperature oxycellulose shows 
only a pale-yellow color, must be accepted with reservation. 
Pure hydrocellulose is stable at these temperatures and much 
more resistant to heat than oxycellulose. 

When treated with alkaline salts, dried, and steamed, oxycel- 
lulose turns brown and becomes about 100 times darker than 
cellulose. Hydrocellulose is ultimately colored less than cellu- 
lose. The chief cause of the yellow coloration of cotton is 
attributed to the presence of /3-oxy cellulose. 7 

Esters of Oxycellulose. — The oxycellulose nitrate prepared 
by Cross and Bevan 8 corresponded with the compound 
Ci 8 H230i6(N0 3 )3 and contained 6.48 per cent of nitrogen. Bull, 9 
however, obtained an ester with 12.2 per cent of nitrogen from 
0-oxy cellulose prepared with nitric acid. The nitrates of cellulose, 
hydrocellulose, and oxycellulose contain about the same amount 
of nitrogen. 10 Their stability in air and their resistance to alkalis 
decrease in the above order. All three nitrates reduce Fehling's 

1 G. Witz and F. Osmund, Bull. soc chim., 45 (1886), 309-315. 

2 W. D. Bancroft, J. Phys. Chem., 19 (1915), 167. 

3 C. Kurz, Z. Farben-und Texlilchem., 1 (1902), 46-47. 

4 E. Fues, G. P. 193909 (1906). 

5 J. Allan, J. Soc. Dyers Colourists, 14 (1898), 7. 

6 E. Justin-Mueller, Bull. soc. chim., 29 (1921), 987-988. 

7 M. Freiberger, Z. angew. Chem., 31 (1918), Ref. 146. 

8 C. F. Cross and E. J. Bevan, J. Chem. Soc, 43 (1883), 22. 

9 B. S. Bull, J. Chem. Soc, 71 (1897), 1096. 

10 L. Vignon, Compt. rend., 126 (1898), 1658. 



276 CHEMISTRY OF CELLULOSE AND WOOD 

solution, the nitrate of oxycellulose showing the greatest aldehydic 
property. 

Knecht 1 found that oxycellulose (by KMn0 4 + H 2 S0 4 ) 
acetylated more slowly than the original cellulose, but eventually 
reached about the same degree of esterifi cation. On nitration, 
a distinct suppression of hydroxyl activity was noted, the 
oxycellulose nitrate containing only 11.5 per cent of nitrogen in 
comparison with 13.4 per cent for the nitrate prepared from the 
original cellulose. 

Oxycellulose, after acetylation, 2 retained its aldehydic proper- 
ties, and apparently contained four acetyl groups. It should not 
be assumed, however, that, as a result, there are three secondary 
alcohol groups in the oxycellulose. 

Oxycelluloses on acetolysis by the method of Skraup and Konig 
gave the same cellobiose octacetate, the yield varying with the 
method of preparing the oxycelluloses. 3 

Cellobiose 

Octacetate, 

Per Cent 

Cellulose 25 

"Hydralcellulose" (by H 2 2 ) 25 

Oxycellulose (by HN0 3 ) 16 

Oxycellulose (by KCIO, + HC1) 10 

Acid cellulose (by precipitation with acid from Schweizer's 

solution) 7 

Oxycellulose, formed by electrolysis of a potassium chloride 
solution containing cellulose, was more readily acetylated than 
normal cellulose, but the acetate contained less combined acetic 
acid. Acetolysis by Ost's method gave 30 to 31 per cent of 
crystalline cellobiose octacetate. 4 

Detection of Oxycellulose. — Many qualitative methods have 
been proposed for detecting oxycellulose. Most of them are 
based on the property of reduction, hence do not distinguish 
between oxycellulose and hydro cellulose. As the property of 
being dyed directly by basic dyestuffs, such as methylene blue, 
was not sufficiently well denned, it was proposed to use Fehling's 
solution, whereby fibers containing oxycellulose are colored 

1 E. Knecht, J. Soc. Dyers Colourists, 38 (1922), 132. 

2 L. Vignon and F. Gerin, Compt. rend., 131 (1900), 558. 

3 E. R. von Hardt-Stremayr, Monatsh., 28 (1907), 73-78. 

4 R. Oertel, Z. angew. Chem., 26 (1913), 246. 



OXY 'CELLULOSE 277 

red by the cuprous oxide formed. 1 Nessler's reagent gives a 
gray color, 2 and Harrison's solution containing 1 per cent of 
silver nitrate, 4 per cent of sodium thiosulphate, and 4 per cent 
of sodium hydroxide, a gray to black color. 3 

Several methods depend upon the behavior of oxycellulose 
with dyes. Cotton spotted with oxycellulose is rinsed in acid, 
then many times with water, and dyed a deep shade with congo 
red. The cotton is then placed in sufficient acid to develop a 
blue color. On washing carefully with a limited amount of 
water until the ordinary cotton has a good red shade, the spots 
of oxycellulose will appear black on the red ground color. 4 

Oxidized cellulose develops a brown color proportional to its 
degree of oxidation when treated for 10 minutes with 750 cubic 
centimeters of an ice-cold solution containing 0.01 gram-molecule 
of tetrazotized benzidine and 80 cubic centimeters of normal 
sodium hydroxide. The reagent reacts to only a slight extent 
with hydrocellulose and not at all with normal bleached cotton, 
mercerized cotton, or with cotton so tendered with sulphuric 
acid as to have an increased affinity for methylene blue. 5 

Owing to its reducing properties, highly oxidized cellulose is 
able to fix flaventhrene directly, on boiling with 2N sodium 
hydroxide, by conversion of the dye to a blue vat. 6 On exposure 
to air, the blue color changes to yellow. Oxycellulose when 
boiled with direct cotton dyes, such as rosinduline 2B, and sodium 
hydroxide, destroys a large amount of the dye. Prolonged boiling 
removes the oxycellulose completely. The amount of dye 
destroyed can be determined by titration. Direct colors of the 
primuline type, that are unaffected by reducing agents, dye oxy- 
cellulose immediately. 7 

Oxycellulose has only a slight affinity for leucomethylene blue 
and may be distinguished in this way from cotton treated with 

1 M. Philip, Z. offentl. Chem., 6 (1900), 524; L. L. Lloyd, J. Soc. Dyers 
Colourists, 26 (1910), 273. 

2 H. Ditz, /. prakt. Chem., 78 (1908), 343. 

3 W. Harrison, J, Soc. Dyers Colourists, 28 (1912), 361. 

4 A. B. Knaggs, J. Soc. Dyers Colourists, 24 (1908), 112. 

5 A. E. Everest and A. J. Hall, J. Soc. Dyers Colourists, 39 (1923), 
47-49. 

6 R. Scholl, Ber., 44 (1911), 1312. 

7 E. Knecht, J. Soc. Dyers Colourists, 37 (1921), 76-77. 



278 CHEMISTRY OF CELLULOSE AND WOOD 

dilute sulphuric acid, which has a marked affinity for it. 1 The 
affinity of the oxy cellulose for a direct dye, such as diamine sky 
blue, may be restored by boiling it with dilute alkali. 

The only qualitative tests characteristic of oxycellulose are 
those based on the activity of the carboxyl group. Oxycellulose 
gave a color reaction very similar to glucuronic acid by Tollens' 
napthoresorcinol method; it differed, however, in that the ether 
solution was colored reddish violet and no spectral bands were 
observed. 2 The test for glucuronic acid is carried out as follows : 3 
Place in a small test tube about 20 milligrams of the substance 
to be examined, 5 to 6 cubic centimeters of water, 0.5 to 1.0 cubic 
centimeter of a 1 per cent solution of napthoresorcinol in alcohol, 
and 5 to 6 cubic centimeters of hydrochloric acid of sp. gr. 1.19. 
The mixture is warmed slowly to boiling, boiled for a minute, 
let stand 4 minutes, and cooled in running water. An equal 
volume of ether is then added and the contents shaken. The 
ether layer is colored blue or violet and shows a dark spectral 
band near the D line. 

The reaction is not specific for glucuronic acid, but is given by 
a large number of substances containing aldehyde-carboxyl or 
keto-carboxyl groups. 4 

A useful test is that of Ditz 5 as amplified by Schwalbe and 
Becker. 6 In the presence of sodium chloride, the carboxyl group 
in oxycellulose turns methyl orange wine red. The cellulose to 
be examined is suspended in water; addition of a drop of methyl 
orange colors the liquid yellow or reddish yellow. On introduc- 
ing a few cubic centimeters of a concentrated solution of sodium 
chloride, the color, in the presence of oxycellulose, changes to wine 
red; with other forms of cellulose there is practically no change. 
The method may be rendered quantitative by titration with 
standard alkali. Even here the distinction is not as sharp as 
might be desired. 

1 E. Knecht and F. P. Thompson, J. Soc. Dyers Colourists, 37 (1921), 
270-277. 

2 E. Heuser and F. Stockigt, Cellulosechemie, 3 (1922), 70. 

3 B. Tollens, Ber., 41 (1908), 1788; of. Ibid., 1783. 

4 J. A. Mandel and C. Neuberg, Biochem. Z., 13 (1908), 148; C. Neu- 
berg, Ibid., 24 (1910), 436. 

5 H. Ditz, Chem. Ztg., 31 (1907), 844. 

6 C. G. Schwalbe and E. Becker, Ber., 54 (1921), 545. 



OXYCELLULOSE 



279 



Material 



Corrected 
copper 
number 



Cubic 
centi- 
meters 
O.OIN 
alkali* 



Color with 

methyl 

orange and 

NaCl 



1. Oxycellulose (by bleaching pow- 

der) 

2. Nitrating cotton 

3. Wood pulp oxycellulose (by 

bleaching powder) 

4. Wood pulp (for nitration) 

5. Oxycellulose (by KMn0 4 ) 

6. Oxycellulose (by H 2 2 ) 

7. Hydrocellulose (Girard) 

8. Wood pulp, Ritter-Kellner, un- 

bleached 

9. Wood pulp, normal bleach 

10. Wood pulp, overbleached 



10.99 


41.4 


0.28 


1.8 


33.22 


33.4 


1.00 


4.1 


8.03 


27.7 


5.80 


10.4 


3.64 


4.6 


1.14 


7.0 


2.14 


7.1 


3.85 


14.9 



Very strong 

rose 
Yellow 

Rose 
Yellow 
Very strong 

rose 
Strong rose 
Yellowish 

brown 

Yellow 

Yellow 
Yellow 



* For 1 gram of material. Figures in this column are corrected for the alkalinity of the 
ash. 

The "acid number" of Vieweg 1 shows that oxycellulose is 
more acid than hydrocellulose. This is obtained by boiling a 
weighed portion of the cellulose (3 grams) for 15 minutes with 
a measured quantity (200 cubic centimeters) of 0.5iV sodium 
hydroxide and determining by titration the loss in alkalinity of a 
withdrawn portion of the liquor. 

Formation of Furfural. — Oxycellulose on distillation with acids 
gives furfural. 2 Starch and sugars after oxidation with chromic 
acid gave as high as 11 per cent. Owing to the difficulty of 
isolating pentoses or obtaining a pentose reaction from many 
celluloses that gave furfural on acid distillation, Cross and Bevan 
assumed that they were natural oxycelluloses and contained 
"furfuroids." The best experimental evidence indicates that 
furfural is obtainable in quantity only from pentosans, under 



1 W. Vieweg, Papier-Ztg., 34 (1909), 1352. 

2 C. F. Cross, E. J. Bevan, and C. Beadle, Ber., 26 (1893), 2520; 27 
(1894), 1061, 1456. 



280 



CHEMISTRY OF CELLULOSE AND WOOD 



which subject the "natural oxy celluloses" are discussed in more 
detail. 

In general, the more drastic the oxidation the smaller is the 
yield of oxycellulose and the greater the yield of furfural obtain- 
able. The statement 1 that the acid distillate from oxycellulose 
gives a brownish-black methylfurfural phloroglucide completely 
soluble in cold 80 per cent alcohol, hence free from furfural phloro- 
glucide, is open to question. It is much more probable, from the 
work of Cunningham and Doree, 2 that hydroxymethylfurfural 
is formed. This aldehyde gives a red phloroglucide and methyl- 
furfural a red-brown one. Oxycellulose obtained by the treat- 
ment of mercerized cotton with ozone gave 1.11 per cent of 
furfural and 0.66 per cent of hydroxymethylfurfural based on 



Furfural from Oxycellulose 



Prepared with 



Furfural, 


per 


cent 


4.1 


-8.2 


2.11 


-2.09 


3.50-1.82 


1.80-1.60 


1 


79 


2 


6 


7 





2.3 


-3.2 


1.4- 


1.9 


1 


09 


1 


09 


1 


7 





80 


3 


05 


1 


90 


3 


89 


2 


1 





65 


1 


1 



Author 



Reference 



CrOs 

KClOs + HC1 

Cr0 3 

Bleached cotton . . . 

Alkaline NaOCl... 
Ca(OCl) 2 

HNO3 (wood) 

HNO3 (cotton) 

Br 2 + CaC0 3 

Ca(OCl) 2 

KMn0 4 + H2SO4. 
Electrolysis of KC1 

KCIO3 + HC1 

CrOs + H2SO, . . . . 
KMn0 4 + H2SO4. 

Cr0 3 

Br 2 + CaCOs 

H2O2 

Ozone 



Cross, Bevan, and 

Beadle 
Vignon 

Vignon 

Vignon 

Vignon 

Teomp de Haas and 

TOLLENS 

Von Faber and Tol- 

LENS 

Von Faber and Tol- 

lens 
Von Faber and Tol- 

lens 
Nastjukoff 
Nastjukoff 
Oertel 
Zanotti 
Zanotti 
Zanotti 

Heuser and Stockigt 
Heuser and Stockigt 
Heuser and Stockigt 
Cunningham and 

Doree 



Ber., 26 (1893), 2522 



1356 



1356 



1356 



Compt. rend., 126 (1898) 

128 (1899), 1038 
Compt. rend., 126 (1898) 

128 (1899), 1038 
Compt. rend., 126 (1898) 

128 (1899), 1038 
Compt. rend., 128 (1899), 1038 
Ann., 286 (1895), 296 

Ber., 32 (1899), 2589 

Ber., 32 (1899), 2589 

Ber., 32 (1899 , 2589 

Ber., 33 (1900), 2237 
Ber., 33 (1900), 2237 
Z. angew. Chem., 26 (1913), 246 
Chem. Zentr., 70, I (1899), 1209 
Chem. Zentr., 70, I (1899), 1209 
Chem. Zentr., 70, I (1899), 1209 
Cellulosechemie, 3 (1922), 61 
Cellulosechemie, 3 (1922), 61 
Cellulosechemie, 3 (1922), 61 
J. Chem. Soc, 101 (1912), 500 



1 F. Lenze, B. Pleus, and J. Muller, J. prakt. Chem., 101 (1920), 250. 

2 M. Cunningham and C. Dor£e, Biochem. J., 8 (1914), 438. 



OXYCELLULOSE 281 

the solubility of the mixed phloroglucides in alcohol. Hydroxy- 
methylfurfural is formed more slowly and appears only in small 
amount as an impurity towards the end of the acid distillation. 1 

When oxidized cellulose is extracted with alkali, the portion 
precipitated by acidification of the alkaline filtrate gives an 
appreciably higher percentage of furfural than does the part 
insoluble in alkali or that not precipitated from the alkaline 
extract. 2 This indicates that the formation of furfural is directly 
linked up with the acid portion of the product. 

The furfural distillate from oxycellulose prepared with chromic 
acid contains 13.8 per cent of substances behaving towards 
phloroglucinol like furfural. They are destroyed by redistilla- 
tion of the hydrochloric acid distillate. 3 

a-, (5-, and y-Oxycelluloses and Their Salts. — The oxycelluloses 
prepared with bleaching powder and nitric acid have been called 
a- and jS-oxycellulose, respectively. 4 NastjukofT 5 has described 
a third form, 7-oxy cellulose. The latter is obtained in yields 
of 60 to 80 per cent by heating the a-oxycelluloses, prepared 
with bleaching powder and potassium permanganate, with 5 
per cent sulphuric acid for 3 and 1 hours, respectively, washing, 
then heating with 10 per cent sodium carbonate solution for 10 
to 30 minutes at 70 to 100°. 7-Oxycellulose gives aqueous 
opalescent solutions that pass through filter paper and remain 
unaltered on standing or heating. Solutions of a concentration 
of 5 to 10 per cent resemble glycerine or viscose. Precipitation 
with acid renders it insoluble in water, but it becomes soluble 
again on treatment with sodium carbonate. It reduces Fehling's 
solution and gives a yellow hydrazone. The soluble form is the 
sodium salt of an acid giving on evaporation lustrous films that 
may be easily detached from glass. After drying at 80 to 100°, 
the salts of /3-oxycellulose show decreased solubility in water, 
while those from a-oxy cellulose are not affected. 

When precipitated from alkaline solution with a salt or 
alcohol, /3-oxycellulose was almost free from ash, hence contained 

1 E. Heuser and F. Stockigt, Cellulosechemie, 3 (1922), 69. 
2 L. Vignon, Compl. rend., 126 (1898), 1357. 

3 G. S. Fraps, Am. Chem. J. 25 (1901), 505. 

4 C. F. Cross and E. J. Bevan, J. Sdc. Chem. Ind., 3 (1884), 206. 

5 A. Nastjukoff, Ber., 34 (1901), 719, 3589. 



282 CHEMISTRY OF CELLULOSE AND WOOD 

little if any combined alkali. 1 /3-Oxy cellulose gave a barium 
salt containing about 5 per cent of barium, while the correspond- 
ing salt of 7-oxycellulose, prepared with bleaching powder, con- 
tained only 1 per cent of barium. 2 

Oxidizing agents, according to Piest, 3 convert cellulose into a- 
and /3-oxycellulose, the former being insoluble and the latter 
soluble in 5 per cent sodium hydroxide in the cold. The a- 
oxycellulose had all the properties of normal cellulose and should 
more correctly be considered as such. Cotton on boiling with 
30 per cent sodium hydroxide dissolves partially, the portion 
going into solution being an acid cellulose. 

Nastjukoff 4 has characterized the three forms of oxy cellulose 
as follows: 

1. a-Oxy cellulose (Witz), insoluble in ammonia and dilute 
alkalis. 

2. /3-Oxycellulose, soluble in ammonia. 

3. T-Oxy cellulose, soluble in ammonia and the fixed alkalis, 
and in water as well. 

According to Bancroft, 5 there is no justification for assuming 
the existence of three oxycelluloses. It is most probable that 
a-oxycellulose is unchanged cellulose and that oxycellulose 
consists of cellulose contaminated with various oxidation prod- 
ucts. Regardless of the oxidizing agents employed, the result- 
ing products have the following characteristics: 

1. No change in color when heated in air to 100°. 

2. They form a gelatinous mass when warmed with water. 

3. They are partially soluble in KOH or NaOH with a yellow 
color. 

4. They dissolve readily in water, after precipitation from 
alkaline solution with alcohol or acid, followed by dialysis. 

5. They reduce Fehling's solution, but this property is lost by 
too long heating. 

One must agree with Bancroft that no pure oxycellulose has 
been prepared as yet. As will be shown elsewhere, the structure 
of cellulose renders this end extremely difficult of attainment. 

1 C. F. Cross and E. J. Bevan, /. Chem. Soc, 43 (1883), 22. 

2 A. Nastjukoff, Ber., 34 (1901), 3590. 

3 C. Piest, Z. angew. Chem., 26 (1913), 27. 

4 A. Nastjukoff, Ber., 34 (1901), 720, 722, 3590. 

6 W. D. Bancroft, J. Phys. Chem., 19 (1915), 159-168. 



OXYCELLULOSE 283 

Composition of Oxycellulose. — Nastjukoff 1 believed that the 
oxy cellulose of Witz consisted of a mixture of cellulose and 
oxycellulose. On heating with strong caustic soda by Lange's 
method, 2 there remained a residue of 64.5 to 69.4 per cent of 
cellulose. The residue from the treatment of oxycellulose with 
alcoholic sodium hydroxide is a hydrated cellulose. 3 

According to Tollens, 4 oxy celluloses are mixtures, or, rather, 
chemical combinations, of cellulose with a more highly oxygen- 
ated body, C 6 Hi O 6 or C 6 H 8 6 , termed "celloxin." The latter 
is the true oxycellulose not yet isolated. When 30 grams of 
oxycellulose were boiled with lime water, there remained 23.5 
grams of a material having all the properties of a pure cellulose. 5 
As a result of fusing oxycellulose (by HC1 and KC10 3 ) with 
potassium hydroxide at 180°, Vignon 6 concluded that it contained 
75 per cent of oxycellulose and 25 per cent of cellulose. Strong 
solutions of the alkalis decompose oxycellulose into cellulose 
and a soluble acid. 7 Schwalbe and Becker 8 remove the degrada- 
tion products from cellulose by boiling with the alkaline earths. 
In this way, an oxycellulose prepared with bleaching powder left 
a 60 per cent residue consisting of practically pure cellulose. 

Oxycellulose consists of a mixture of cellulose, hydrocellulose, 
and their oxidation products. 9 According to Harrison, it con- 
sists of adsorption compounds of peptized cellulose and reducing 
substances. 10 

Nature of Oxycellulose. — There is much evidence for the 
presence of aldehyde groups in oxycellulose. Witz 11 observed 
that oxycellulose as well as its alkaline solution had a strong 
reducing action on Fehling's solution, while hydrocellulose had 

1 A. Nastjukoff, Bull. Mulhouse, 62 (1892), 493-510. 

2 G. Lange, Z. physiol. Chem., 14 (1890), 286. 

3 A. Nastjukoff, Ber., 33 (1900), 2237. 

4 O. von Faber and B. Tollens, Ber., 32 (1899), 2589. 

5 J. J. Murumow, J. Sack, and B. Tollens, Ber., 34 (1901), 1431. 

6 L. Vignon, Compt. rend., 125 (1897), 449. 

7 L. Vignon, Compt. rend., 131 (1900), 708. 

8 C. G. Schwalbe and E. Becker, J. prakt. Chem., 100 (1919), 19. 

9 V. Zanotti, Chem. Zentr., 70, I (1899), 1209. 

10 W. Harrison, J. Soc. Dyers Colourists, 28 (1912), 360. 

11 G. Witz, Bull. Rouen, 11 (1883), 219; C. F. Cross and E. J. Bevan, 
Chem. News, 64 (1891), 63. 



284 



CHEMISTRY OF CELLULOSE AND WOOD 



<A 



00 
CN OS 
CN 00 



CO CO 
00 O to 
00 OS § 

^ C- co 



_ Os OS OS 

S o o o co 

S N <M IN O) 

« r-H ,-H tH O 



OS 



-^ ^ P3 

3 r^ 



92 os os 

OS as OS 
00 GO GO 



^ tF <N 

OS OS Os 

lO iQ lO 

(M (N (N 



Os Os OS 
OS Os OS 
00 00 GO 



N 03 C3 

M CO W 



cq cq oq 



N H H H 



os ^ 
os t- 

00 OS 

H00 H 
O 

to 

,. "OS 
" W OS 

*<83 



Tj ^j Sjj Sy • 

>5j ESj tS] ES3 g 

~ to 



g s s 

to OS to 



3 rf; -e 



cq 



a 
O ? 



w 



o 



< 
> 

pq 

!=! 
o3 

02 
S3 02 
H O 



02 02 CO 

fc £ fc 
B B S 



o S 

fc ft 



Eh H H 

TJ TJ "O" 

fl fl fl 

c3 o3 Cj 

tf as « 

B B B 

g sa ^ 

P=H P=H P=H 

fc fc ^ 

o o o 



Ei U 

o o 



> h qfe>>>^sitsiNPQm 



B 

M £ 

% g 

M 02 



oooooooo^ 
c5c5doduoc5c5 



OOOOOOCOCOCOCN 
COC01>T]HcOC01>t^ 


• ^ CM <N i-h O 

• N lO © 00 O 


Oi-iOOt-hOOt-i 


■ OS O O rH O 


OSOOOOSiOiOOSCO 

ffilNOJHOOOOH 


•OlNtDN 
. CO to tO CM CM 



lOiOiOtOOCOiOO 



CD CO CO CO CD 



O 



o3 s 



-d t3 



"ti 'd q 



-f3 += -(-= 



I O O HH 

.£ ^ _L "• 

03 r^ 



§ :|3§S 



5 Jzj M £ £ jzj fe m « M u M g (& 



OXY CELLULOSE 



285 



a feeble one. In addition to reducing Fehling's solution, it 
colors SchifPs reagent easily and intensely. 1 

Oxycellulose contains aldehydic or ketonic oxygen, since it 
reacts with phenylhydrazine. 2 The osazones prepared from the 
oxycelluloses from cotton, flax, hemp, and ramie were considered 
to be identical. They contained 1.58 to 1.69 per cent of nitro- 
gen. 3 Vignon 4 found that the greater the degree of oxidation, 
the greater the amount of nitrogen fixed by the oxycellulose 
when boiled with phenylhydrazine. Acid oxidizing agents are 
preferable, since alkaline ones largely destroy the reducing prod- 
ucts as formed. 



Oxycellulose prepared with 


Nitrogen 

fixed, 
per cent 


Corresponding 

phenylhydrazine, 

per cent 


KC10 3 + HC1 


2.06 

0.87 

1.82 
2.00 
2.20 


7.94 


NaOCl (alkaline; boiling) 


3.37 


K 2 Cr 2 7 + H0SO4: 

a. 48 hours in the cold 


7.03 


b. 120 hours in the cold 


7.71 


c. 1 hour, boiling 


8.48 







Oxycellulose, like the aldehyde sugars, gives color reactions 
with phenols in the presence of sulphuric acid: with a-napthol, 
violet; with /3-napthol, brown; and with thymol, rose. 5 

The various oxycelluloses show marked differences in the yield 
of furfural, resistance to alkalis, acidity, and reducing properties. 6 
In one class of oxycelluloses the acid, in the other the aldehyde, 
character predominates. In hydrocellulose the acid property 
is almost entirely absent. Denitrated wood cellulose and Char- 
donnet artificial silk show the combined properties of oxycellulose 

1 E. R. Flint and B. Tollens, Ann., 272 (1893), 289; L. Vignon, Compt. 
rend., 125 (1897), 448. 

2 A. Nastjukoff, Bull. Mulhouse, 62 (1892), 510; A. Scheurer and 
E. N6LTING, Ibid., p. 492. 

3 L. Vignon, Compt. rend., 131 (1900), 558. 
4 L. Vignon, Compt. rend., 128 (1899), 1038. 

5 E. Jandrier, Compt. rend., 128 (1899), 1407. 

6 C. G. Schwalbe and E. Becker, Zellstoff u. Papier, 1 (1921), 100-103, 
135-139. 



286 CHEMISTRY OF CELLULOSE AND WOOD 

and hydrated cellulose, the former showing chiefly the character 
of an oxy cellulose, the latter that of hydrated cellulose. 

Oxycellulose when heated with lime water gave isosaccharinic 
acid and dihydroxybutyric acid. 1 No very definite conclusions 
can be drawn from this reaction, however, since several carbo- 
hydrates, including hydrocellulose, give these products. Tollens 2 
concluded that oxycellulose contained a carboxyl group and an 
aldehyde or ketone group, since it formed osazones and reduced 
Fehling's solution. 

Glucose, on treatment with lime water, gives metasaccharinic 
acid, CH 2 OH.[CHOH] 2 .CH 2 .CHOH.COOH, and not isosac- 

X)H 2 OH 
charinic acid, CH 2 OH.CHOH.CH 2 .cA)H . Since iso- 

X COOH 
saccharinic acid has apparently been obtained only from lactose 
and maltose, Pringsheim 3 concludes that the formation of this 
acid is limited to a glucose group having a glucosidal linkage. If 
cellobiose should be shown to give isosaccharinic and not meta- 
saccharinic acid, it would be certain proof that the isosaccharinic 
acid was formed from the oxycellulose or hydrocellulose before 
the rupture of the glucosidal linkage at the aldehyde group of 
the glucose residues. He believes that oxycelluloses consist 
of a mixture of anhydrocellobiose and cellobionic acid held 
together by secondary valencies. Hintikka 4 has found that 
cellobiose does give isosaccharinic acid with lime water. 

Vignon 5 found that the apparent acidity of oxycellulose was 
much greater than for cellulose. On immersing these substances 
in normal potassium hydroxide, the heats liberated for 100 grams 
of material were 1.30 and 0.74 calories respectively. When 
oxycellulose nitrate was dissolved in potassium hydroxide, 
oxy pyruvic acid was isolated as the osazone from the decomposi- 
tion products. 6 Since this compound had been previously 
obtained from cellulose nitrate by Will, 7 Vignon concluded that 

1 O. von Faber and B. Tollens, Ber., 32 (1899), 2596. 

2 B. Tollens, Ber., 34 (1901), 1435. 

3 H. Pringsheim, Cellulosechemie, 2 (1921), 61. 

4 S. V. Hintikka, C. A., 17 (1923), 3486. 
5 L. Vignon, Compt. rend., 125 (1897), 448. 
6 L. Vignon, Compt. rend., 127 (1898), 872. 
7 W. Will, Ber., 24 (1891), 400. 



OXYCELLULOSE 287 

oxy cellulose might be formed during nitration. In his opinion, 
the essential group in oxycellulose is OHC.(CHOH) 3 .CH — CO, 

V 

this being at once an aldehyde, alcohol, and lactone of an acid, 
C 6 Hi O 7 . The formation of oxypyruvic acid and furfural 
takes place as follows : 

C 6 H 10 O 7 + 2 = 2 CHOH.CO.COOH + H 2 0. 

Oxypyruvic Acid 
C 6 H 10 O 7 = C5H4O2 + 3H 2 + C0 2 . 
Furfural 

Berl and Fodor 1 support the view that cellulose nitrates are 
really esters of oxycellulose. The latter contain glucosidohexonic 
acids, Tollen's "celloxin" being the lactone form of the hexonic 
acid, cellonic acid. Cellulose nitrate, 2 on hydrolysis with a 30 
per cent solution of alcoholic potassium hydroxide, gave oxalic 
and pyruvic acids but no saccharic acid; with 10 per cent aque- 
ous potassium hydroxide, oxalic, malic, glycollic, trihydroxy- 
glutaric, dihydroxybutyric, malonic, and tartronic acids were 
obtained. 

The lactone of glucuronic acid, C 6 H 8 6 , gives, in addition to 
furfural, 25 per cent of carbon dioxide, the latter being suitable 
for its quantitative determination. 3 

,- O , 



CO.CHOH.CH.CH.CHOH.CHOH = C 5 H 4 2 + C0 2 + 2H 2 0. 
I n __ I Furfural 

Glucuronic acid lactone 

It was shown by Heuser and Stockigt 4 that oxycellulose 
actually contains a group related to, if not identical with, glucu- 
ronic acid. This explains the formation of furfural and carbon 
dioxide when oxycellulose is distilled with 12 per cent hydro- 
chloric acid. Oxycellulose prepared with chromic acid gives the 

1 E. Berl and A. Fodor, Z. Schiess-u. Sprengstoff., 5 (1910), 254, 269. 
2 E. Berl and A. Fodor, Z. Schiess-u. Sprengstoff., 5 (1910), 296, 313. 

3 K. U. Lefevre and B. Tollens, Ber., 40 (1907), 4517. 

4 E. Heuser and F. Stockigt, Cellulosechemie, 3 (1922), 61. 



288 CHEMISTRY OF CELLULOSE AND WOOD 

highest yields of furfural and carbon dioxide. The yields of 
carbon dioxide from various carbohydrates were: 

co 2 

Per Cent 

Glucose . 04 

Pure cotton . 026 

Hydrocellulose . 04 

Sulphite pulp (unbleached) . 08 

Sulphite pulp (bleached) . 24 

Soda pulp (bleached) . 22 

Oxycellulose (by chromic acid) 1 . 53 

Glucuronic acid lactone 24 . 84 

Even though glucose on prolonged heating with dilute mineral 
acids gives from 1.5 to 2.0 per cent of carbon dioxide, 1 the forma- 
tion of furfural and carbon dioxide under the conditions specified 
must be considered as characteristic of oxycellulose. 

It was found that by heating the crude oxycellulose with 1 
per cent sulphuric acid under pressure, it was possible to remove 
the small portion giving furfural and carbon dioxide. The 
residue was hydrocellulose. The hydrolyzed portion gave an 
impure, unstable barium salt related to glucuronic acid. A 
reddish- violet color was obtained by the /3-napthoresorcinol test, 
but the ether solution showed no spectral bands. The oxycellu- 
lose, on hydrolysis with strong hydrochloric acid, gave only 10 
per cent less glucose than pure cellulose. 2 

Tollens 3 has suggested that the furfural obtained from oxycel- 
lulose may be derived from certain groupings related to glucosone, 
CH 2 OH.[CHOH] 3 .CO.CHO, which also gives furfural. 4 

In the light of our present knowledge of the constitution of 
cellulose, the degradation of cellulose to glucuronic acid, with the 
intermediate formation of oxycellulose, must take place in several 
stages. In the glucose anhydride units shown below, the dashes 
represent oxygen linkages. The first step involves the oxidation 
of the primary alcohol group A to the aldehyde B ; the next being 
the oxidation of the latter to the carboxyl group C. Hydrolysis 
then gives glucuronic acid D. 

1 M. Berthelot and G. Andre, Compt. rend., 123 (1896), 567. 

2 E. Heuser and F. Stockigt, I.e., 64. 

3 B. Tollens, Ann., 286 (1895), 301. 

4 E. Fischer, Ber,, 22 (1889), 93. 



OX YCELL ULOSE 289 

A B C D 

CH 2 OH CHO COOH COOH 

I I I I 

CH-- CH-- CH-- CHOH 

! 2 I 2 I H 2 I 

--CH > --CH > --CH > CHOH 

I I I I 

[CHOH] 2 [CHOH] 2 [CHOH] 2 [CHOH] 2 

I I I I 

--CH-- --GH-- --CH-- CHO 

F F F Glucuronic acid 

The grouping CF would represent the true "oxj^cellulose," 
however inappropriate the nomenclature of the latter. It is 
highly improbable that more than a small amount of CF could 
be formed, since all known oxidizing agents appear to affect 
cellulose profoundly. The alkaline agents rapidly produce 
total disintegration, while oxidation in the presence of acids 
produces simultaneous hydrolysis. Even hydrogen peroxide 
hydrolyzes the cellulose molecule. 

The reducing action of oxycellulose may be due to both the 
aldehydic group B formed by oxidation and to that resulting 
from hydrolysis at the glucosidal linkage at F. According to the 
course of the reaction, an oxycellulose may be pronouncedly 
aldehydic or acid, but it will have both properties. It will, 
obviously, be extremely difficult to control the oxidation so as 
to obtain a uniform product. 



CHAPTER IX 
THE ACTION OF ACIDS ON CELLULOSE 

Acids decompose cellulose according to their nature and 
strength. Dilute acids at the boiling point, or when allowed to 
dry on the fibers, produce chiefly hydrocellulose. Under more 
drastic conditions, hydrolysis proceeds to dextrins and the simple 
sugars. Oxidizing acids give both hydrocellulose and oxycellu- 
lose, and finally the simple aliphatic acids. Under certain 
conditions, acids may cause hydrolysis to the simple sugars, 
followed by dehydration to furfural and hydroxymethylfurfural. 

Hydrocellulose. — Hydrocellulose is an indefinite product 
consisting of small amounts of bodies of the nature of dextrins 
associated with unchanged cellulose. Fibers which have been 
converted to hydrocellulose are characterized by brittleness and 
the presence of aldehyde groups. The absence of carboxyl 
groups distinguishes it from oxy cellulose. 

The effect of dilute mineral acids in decreasing the strength 
of linen was pointed out by Kolb 1 in 1868. A few years later 
Girard 2 described a modification of cellulose, obtained by the 
action of mineral acids of moderate concentration, and called by 
him hydrocellulose. The latter was obtained by immersing 
cotton in sulphuric acid of 45° Be. (55 per cent) for 12 hours, or 
by impregnating the fiber with very dilute acid and heating at 
100°. The product reduced Fehling's solution and dissolved, on 
heating in dilute potassium hydroxide solution, to a strongly 
colored liquid with reducing properties. The friability of paper 
from which bleaching powder had not been completely removed 
was attributed to hydrocellulose formed by the hydrochloric 
and hypochlorous acids set free by the carbon dioxide of the air. 

1 J. Kolb, Bull. Mulhouse, 38 (1868), 922. 

2 A. Girard, Compt. rend., 81 (1875), 1105-1108. 

290 



THE ACTION OF ACIDS ON CELLULOSE 291 

Cotton, linen, wood, and other forms of cellulose are converted 
into hydrocellulose by nitric, sulphuric, and hydrochloric acids. 1 
Phosphoric acid acts with difficulty. Moist hydrogen chloride 
gas forms hydrocellulose in an hour; on heating, in a few minutes. 
Cotton, saturated with a 1 per cent solution of the mineral acids 
and the excess acid removed, when allowed to air dry, is con- 
verted into hydrocellulose at the end of 3 months. Heated to 
60 to 70°, the transformation takes place in a few hours. At 
this temperature an acid of a concentration of 0.001 per cent has a 
noticeable effect. The chlorides of zinc, iron, and aluminum 
at 80° convert cellulose into white hydrocellulose. The conver- 
sion loss is about 10 per cent. 

Hydrocellulose differs from ordinary cellulose in the following 
ways. When heated in a sealed tube at 160° with 10 parts of 1 
per cent potassium hydroxide solution, it is rendered completely 
soluble, while ordinary cellulose is unchanged. On heating in 
a sealed tube with 5 per cent sulphuric acid at 180° for 5 to 6 
hours, cellulose dissolves entirely with a slight coloration and no 
formation of gas, while hydrocellulose gives a carbonaceous 
deposit and a noticeable quantity of gaseous products. Hydro- 
cellulose acetylates readily, being converted into a viscous syrup 
by boiling under a reflux condenser with acetic anhydride; 
cellulose is not affected. Pyroxylins made from hydrocellulose 
are very brittle, the nitration proceeding, however, to the same 
extent as with cotton. The prominent physical characteristics 
of hydrocellulose are friability and adhesive properties. 2 

Cotton exposed to the vapors of hydrochloric acid and nitric 
acid (36° Be.) decreased in strength from 24.3 to 9.6, and from 
24.3 to 18.7 respectively, after exposure for 40 minutes. The 
more pronounced effect of hydrochloric acid is due to its greater 
vapor pressure. 3 

Dilute nitric acid at ordinary temperatures gradually changes 
cellulose to hydrocellulose. 4 

1 A. Girard, Ann. chim. phys., [5] 24 (1881), 337-384; Bull. Soc. d'En- 
courg., [3] 9 (1882), 176-198. 

2 A. Girard, I.e. 

3 A. Scheurer, Bull. Mulhouse, 58 (1888), 363. 

4 C. Haeussermann, Z. angew. Chew,., 23 (1910), 1761; 26 (1913), 456; 
Guichard, Bull. soc. chim., 7 (1892), 559. 



292 CHEMISTRY OF CELLULOSE AND WOOD 

According to Lester, 1 cotton containing 0.01 per cent of hydro- 
chloric acid shows no appreciable "tendering" after drying with 
a hot iron and heating in a water oven for 1.5 hours. Cotton, 
tendered by treatment with dilute sulphuric acid and drying 
without heat, gained in strength by neutralization of the acid 
and by mercerization. 2 Cohen 3 found that the loss in strength 
on tendering corresponded with the increase of the copper 
number. Hydrochloric acid must have a strength of at least 
0.008 per cent and sulphuric acid 0.033 per cent when boiled 
with cotton to produce an increase in its copper number. Cotton 
soaked in the dilute acid in the cold, dried, then heated for 10 
minutes at 120° is affected by hydrochloric acid of 0.002 per cent 
and sulphuric acid of 0.008 per cent. 

The loss of strength of cotton boiled with dilute acids of the 
same normality runs approximately parallel with the relative 
conductivities of the acids. 4 

The tendering of cotton dyed with sulphide colors is due to 
the action of sulphuric acid formed by atmopheric oxidation. 5 

Organic Acids. — Calvert 6 treated linen and cotton fibers with 
2 to 4 per cent solutions of tartaric, citric, and oxalic acids, and 
after air drying, heated them at 80, 100, and 126°. No appreci- 
able loss in strength was obtained except with oxalic acid at 126°. 
Cotton saturated with 5 per cent solutions of oxalic, tartaric, 
and citric acids, and heated at 100°, is completely changed into 
hydrocellulose by the former, while tartaric acid has less action, 
and citric acid the least. 7 

Scheurer 8 studied the effect of dilute solutions of oxalic, lactic, 
tartaric, citric, thiocyanic, orthophosphoric, metaphosphoric, 

1 J. H. Lester, J. Soc. Chem. Ind., 34 (1915), 934-936. 

2 H. Wilkinson, J. Soc. Dyers Colourists, 33 (1917), 148-151. 

3 M. Cohen, /. Soc. Dyers Colourists, 31 (1915), 162-165; cf. W. A. 
Lawrance, Can. Chem. J., 3 (1919), 329-331. 

4 M. Fort and F. Pickles, J. Soc. Dyers Colourists, 31 (1915), 255-260. 

5 J. E. Pilling, J. Soc. Dyers Colourists, 22 (1906), 54-65; G. E. Holden, 
Ibid., 26 (1910), 76-78; E. Vlies, Ibid., 26 (1910), 79-81; W. Ebbers, 
Z. Farben-und Textilchem., 1 (1902), 92; T. R. Appleyard andT. B. Deakin, 
J. Soc. Dyers Colourists, 18 (1902), 128. 

6 F. C. Calvert, Bull. Mulhouse, 29 (1858), 208-213; cf. J. A. Barral 
and Salvetat, Ann. chim. phys., [5] 9 (1876), 129. 

7 A. Girard, I.e.: G. E. Pilkington, J. Soc. Dyers Colourists, 31 (1915), 
149. 

8 A. Scheurer, Bull. Mulhouse, 74 (1904), 211. 



THE ACTION OF ACIDS ON CELLULOSE 293 

pyrophosphoric, and phosphorous acids on cotton under the 
influence of hot air and steam. Of these acids, thiocyanic pro- 
duced the greatest decrease in strength in air at 40 to 50°. 

Under the usual conditions of half-wool dyeing and finishing, 
acetic and formic acid reduce the tensile strength of cotton 
about one-half. 1 

In time, oxalic acid converts cellulose to hydrocellulose even 
at the ordinary temperature. When cellulose containing oxalic 
acid is heated in the dry state, apparently a small portion of the 
acid combines to form an ester. 2 

When heated at 100° with 8 to 16 per cent solutions of oxalic 
acid for 2.5 to 12 hours, viscose yields 97 to 98 per cent of hydro- 
cellulose. 3 The latter has a copper number of 10.47 to 10.83 
and is completely soluble in 8 per cent sodium hydroxide. The 
hydrocellulose obtained from cotton in this way is only partially 
soluble in 10 per cent sodium hydroxide. When heated with a 
5 per cent solution of oxalic acid at 180° for 30 minutes, viscose 
hydrocellulose gave 11.86 per cent of hydroxymethylfurfural, in 
comparison with 5.74 per cent from viscose itself, and 0.81 per 
cent from ordinary cotton. 

Salts. — Salts producing acids by hydrolysis likewise form 
hydrocellulose. Girard 4 observed that the chlorides of zinc, 
iron, and aluminum at 80° convert cellulose into white hydro- 
cellulose. Barral and Salvetat 5 tried various salts as carbonizing 
agents. The fibers were saturated with salt solutions of 5° 
Be. and heated in an oven at 140° for 30 minutes. 

Remarks 

Chlorides of Na, K, NH 4 , Ba, Hg. . No effect 

Chlorides of Ca, Mg Slightly brittle 

Chlorides of Fe, Sn, Zn, Cu Black and brittle 

Nitrates of Pb, Na, K, Ca, Ba, NH 4 No effect 

Nitrates of Mg, Fe, Ag Brittle 

Sulphates of Cu, NH 4 , Mn, Fe, Ca, Mg, Na, K No effect 

Sulphate of Zn Slight 

Sulphates of Sn, Al Pronounced 

1 W. Zanker and O. Mann, Fdrber-Ztg., 27 (1916), 355. 

2 J. F. Briggs, J. Soc. Chem. Ind., 31 (1912), 520-522. 

3 E. Heuser and F. Eisenring, Cellulosechemie, 4 (1923), 25. 

4 A. Girard, Ann. chim. phys., [5] 24 (1881), 337. 

5 J. A. Barral and Salvetat, Ann. chim. phys., [5] 9 (1876), 126-131. 



294 CHEMISTRY OF CELLULOSE AND WOOD 

Preparation.- — According to Ost, 1 a hydrocellulose with con- 
stant properties is best obtained by treating 5 grams of cotton 
with 40 grams of glacial acetic acid and 2 grams of concentrated 
sulphuric acid, and allowing the mixture to stand at room tem- 
perature for 2 days. Jentgen 2 treated 10 grams of cellulose 
with 300 cubic centimeters of glacial acetic acid and 2 cubic 
centimeters of hydrochloric acid, at the boiling point, for 10 to 15 
minutes. When sulphuric acid is substituted for the latter, the 
conversion is rapid, but the hydrocellulose has a tendency to 
gelatinize on washing. 

Hydrocellulose may be obtained by melting ferric chloride, 
FeCl 3 .6H 2 0, at a low heat and gradually adding absorbent cotton 
in an amount such that the resulting solution does not become 
too thick. On pouring into water, a precipitate of finely divided 
hydrocellulose is obtained. 3 It is difficult to free the precipitate 
from iron. 

Tollens 4 prepared hydrocellulose by treating cellulose with 
sulphuric acid of sp. gr. 1.52 to 1.54 (61.6 to 63.4 per cent) for 
10 to 48 hours. Acid of the strength used by Girard (55 per 
cent) was not sufficiently strong. The products, most strongly 
attacked, reduced Fehling's solution slightly and were gelatinized 
by washing with water. Strong sodium hydroxide solution 
also gelatinized the hydrocellulose and partly dissolved it. 
Iodine solution gave only a weak coloration, but zinc chloriodine 
a deep blue. Heating with 3 per cent sulphuric acid on the water 
bath for 8 hours dissolved only a little of the hydrocellulose, 
hence only a small amount of reducing sugar was formed. 

Seidel, 5 by the action of dry hydrogen chloride on sulphite 
cellulose, obtained a friable powder similar to the "acid cellulose" 
of Bumcke and Wolfenstein. 6 It dissolved readily in 10 per cent 
sodium hydroxide, from which solution it was precipitable with 
acids. Evidently, either the cellulose or the hydrogen chloride 
was not dry. Girard found that the dry gas was without action. 

1 H. Ost, Z. angew. Chem., 19 (1906), 994. 

2 H. Jentgen, Z. angew. Chem., 23 (1910), 1543; cf. C. G. Schwalbe, 
Ibid., 438. 

3 Z. Northrup, Abstracts Bad., 3 (1919), 7. 

4 J. J. Murumow, J. Sack, and B. Tollens, Ber., 34 (1901), 1431. 

5 H. Seidel, Chem. Zentr., 71, I (1900), 443. 

6 G. Bumcke and R. Wolfenstein, Ber., 32 (1899), 2493. 



THE ACTION OF ACIDS ON CELLULOSE 295 

When moisture is rigidly excluded, the gas is absorbed by the 
cellulose, but the latter can be recovered practically unchanged 
by subsequent washing with water. 1 Dry hydrogen chloride 
is very slow in its action, but solutions of the gas in anhydrous 
alcohol, ether, and benzene have appreciably greater tendering 
action than aqueous solutions of the same concentration. 2 
Perhaps in this case prolonged contact formed water by dehydra- 
tion of the cellulose. 

Cellulose, when heated at a temperature of 60 to 70° with 
glacial acetic acid containing free chlorine, gives a hydrocellulose 
whose acetyl derivatives are characterized by a high solubility 
in water and alcohol. 3 

A product of considerable interest is the ' 'alkali-soluble cellu- 
lose" of Knoevenagel. 4 It is prepared by the various methods 
resulting in the formation of hydrocellulose, but only a modified 
cellulose can be used. Preferably, viscose cellulose containing 
11.5 per cent of moisture is subjected to the action of hydrogen 
chloride gas for 12 hours. Dry cellulose does not react and the 
duration of the acid treatment depends upon the amount of 
moisture in the cellulose. The product from viscose is completely 
soluble in the cold in 8 per cent caustic soda, while that from 
cellulose regenerated from cellulose acetate is only about one- 
tenth as soluble. From the alkaline solution, acids precipitate 
it quantitatively. 

A suitable modified cellulose was also prepared by heating 
sulphite pulp in an indifferent solvent, such as glycerine and 
xylene, a certain amount of moisture being essential for the 
reaction. It had been found by Glum 5 that cellulose, when 
heated above 100° in stable liquids, such as glycerine and oil, 
is so modified that it is about twice as soluble as ordinary cellu- 
lose in the usual cellulose solvents, the same being true of the 
solubility, in appropriate solvents, of the derivatives of the modi- 
fied cellulose. Sulphite pulp, heated in xylene for 6 hours at 140°, 

1 H. Jentgen, Z. angew. Chem., 23 (1910), 1544. 

2 H. F. Coward, F. C. Wood, and F. L. Barrett, J. Textile Inst., 14 
(1923), 520. 

3 "Fabrik Chemischer Preparate von R. Sthamer," G. P. 123121 (1900). 

4 E. Knoevenagel and H. Busch, Cellulosechemie, 3 (1922), 42-60. 

5 O. Glum & Co., G. P. 217316 (1909). 



296 CHEMISTRY OF CELLULOSE AND WOOD 

gave, after the acid treatment, a product of which 94 per cent 
was soluble in cold 8 per cent caustic soda. 1 

Alkali-soluble cellulose is easily soluble in cellulose solvents, 
but insoluble in ammonia and water. It is stained bluish violet 
by zinc chloriodine and iodine (KI + I) solutions, the coloration 
being only slowly removed by washing with water. It is stable 
at 100°. No coloration is obtained with Schiff's reagent. On 
heating with milk of lime, alkali-soluble cellulose loses from 40 
to 50 per cent in weight, but the copper number of the residue 
(13.0) is practically the same as that of the original (14.0). 

In the opinion of Heuser, 2 Girard's hydrocellulose is a mixture 
of cellulose and hydrocellulose, while KnoevenagePs "alkali- 
soluble cellulose" is the true hydrocellulose. The latter behaves 
as a uniform product and gives uniform derivatives on acetyla- 
tion and methylation. Methylation gives dimethylhydrocellu- 
lose, [Ci 2 Hi606.(OCH 3 )4]2, which is a dimethyl derivative of a 
dimeric anhydrocellobiose. 3 

Hess, 4 who examined KnoevenagePs hydrocellulose in cupram- 
monium solution, states that it is a mixture. The specific 
rotation of the product varied with the time of exposure to 
hydrochloric acid vapors. 

Alkali hydrocellulose of the composition Ci 2 H 2 oOio.2NaOH is 
obtained by dissolving viscose hydrocellulose in strong sodium 
hydroxide and preciptating with alcohol. 5 

It has been maintained by Jentgen 6 that the hydrolysis of 
cellulose, e.g., the formation of hydrocellulose, is produced only 
by acids and salts in their molecular state, and not by the ionized 
portion, as is generally assumed. He heated cellulose in solvents 
in which the catalysts could be finely distributed in the molecular 
state, as in glacial acetic acid, ether, and various esters. Hydrol- 
ysis cannot take place without the presence of water, but the 
catalysts are adsorbed by the cellulose and simply act as contact 

1 E. Knoevenagel and H. Busch, Cellulosechemie, 3 (1922), 42-60. 

2 E. Heuser and W. von Neuenstein, Cellulosechemie, 3 (1922), 89-96, 
101-107. 

3 E. Heuser and G. Jayme, Ber., 56 (1923), 1242-1243. 

4 K. Hess, Ann., 435 (1924), 142. 

5 E. Heuser and F. Eisenring, Cellulosechemie, 4 (1923), 25. 

6 H. Jentgen, Z. angew. Chem., 23 (1910), 1541-1546; 24 (1911), 11-12, 
585-586. 



THE ACTION OF ACIDS ON CELLULOSE 



297 



agents. A 1 per cent aqueous solution of an acid hydrolyzes 
cellulose very slowly, while with a non-dissociating solvent the 
hydrolysis proceeds rapidly. This point of view has been 
protested by Schwalbe. 1 

Properties of Hydrocellulose. — Perhaps as a result of a quali- 
tative test only, Vignon 2 incorrectly concluded that hydrocellulose 
had no action on Fehling's solution. This led him to the further 
assumption that the product obtained by the denitrification of 
cellulose nitrate with ammonium sulphide was either cellulose or 
hydrocellulose, since it had no reducing action. 3 

Hydrocelluloses, according to Schwalbe, 4 have a high copper 
number, a low hygroscopic moisture content, a weak affinity 
for basic dyes, and a considerable solubility in the caustic 
alkalis. They are stained blue by zinc chloriodine, but the color 
is readily removed by washing with water. Some properties 



Hydrocellulose from 


Copper 
number 


Hygroscopic 

water, 

per cent 


1. Cotton satin dissolved in concentrated H 2 S0 4 
and precipitated with water 


7.9 
3.9 
4.0 
5.6 
5.6 
5.2 
6.2 
8.7 
8.8 


5.3 


2. Cotton satin with H 2 S0 4 of 45° Be 

3. Cotton satin with HC1 gas 


6.3 
3.8 


4. Filter paper with HC1 gas 


1.2 


5. Cotton satin with 3 per cent H2SO4 


3.6 


6. Absorbent cotton with 3 per cent H2SO4 

7. Filter paper with 3 per cent H2SO4 


3 8 


8. Parchment paper with 3 per cent H2SO4 

9. Mercerized cotton with 3 per cent H2SO4 


6.0 
6.3 



of the hydrocelluloses prepared by him are given above. It will 
be noted that the copper numbers vary from 4.0 to 8.8; Bay 5 
obtained 3.06 to 4.69 for hydrocelluloses prepared by Girard's 
method, using 3 per cent sulphuric acid, while Qvist 6 gives 4.3 

1 C. G. Schwalbe, Z. angew. Chem., 23 (1910), 2030; 24 (1911), 12. 

2 L. Vignon, Compt. rend., 131 (1900), 708. 
3 L. Vignon, Compt. rend., 131 (1900), 530. 

4 C. G. Schwalbe, Z. angew. Chem., 20 (1907), 2166; Ber., 40 (1907), 4523. 

5 C. Bay, "Zur Kenntnis der Hydro-, Oxy-, Hydral-, und Acidcellulosen," 
Diss. Giessen (1913), p. 32. 

6 W. Qvist, Papier-Fabr., 17 (1919), 820. 



298 



CHEMISTRY OF CELLULOSE AND WOOD 



to 5.1 for hydrocelluloses from sulphite pulp. Knoevenagel's 
hydrocellulose 1 had a copper number of about 12.5, while Heuser 2 
obtained 10.47 to 10.83 for a similar product. 

A (8-series of hydrocelluloses, obtained by the aging of alkali 
cellulose, has been described by Cross and Bevan. 3 The "hydro- 
cellulose" formed a microscopically fine powder. It had high 



Hydrocellulose 


a-Series 

by action of 

acids . 


0-Series 

by action of 

alkalis 


Hygroscopic moisture 


3.5-5.0 per cent 
Reducing 


8-11 per cent 
Non-reducing 


Action of Fehling's solution 





hygroscopicity, was without reducing action, and was compara- 
tively stable towards certain esterifying processes. Apparently, 
the only reason for classifying the compound as a hydrocellulose 
was its state of division. This is insufficient. 

The hygroscopic moisture content of hydrocellulose varies 
with the method of conversion, the raw material, and the 
humidity of the atmosphere. Schwalbe found 1.2 to 6.3 per 
cent; Ost and Westoff, 4 4.30 to 4.75 per cent. Absolutely dry 
hydrocelluloses prepared from sulphite cellulose, after standing 
under a bell jar in an atmosphere saturated with moisture, 
contained 9.24 to 11.87 per cent of water. 5 Knoevenagel 6 found 
8.8 to 9.6 per cent hygroscopic moisture for his product in com- 
parison with 4.15 per cent for a hydrocellulose prepared by 
Girard's method. 

Effect of Heat. — Girard found that his hydrocellulose when 
heated in the presence of air turned yellow even at 50°; at 100° 
the decomposition was rapid. The colored product formed 
could be removed by washing with water. His hydrocellulose 
evidently contained traces of acid, since properly purified prod- 



1 E. Knoevenagel, I.e. 

2 F. Heuser and F. Eisenrtng, Cellulosechemie, 4 (1923), 25. 

3 C. F. Cross and E. J. Bevan, Chem. Ztg., 33 (1909), 368. 

4 H. Ost and F. Westhoff, Chem. Ztg., 33 (1909), 197. 

5 W. Qvist, I.e. 

6 E. Knoevenagel, I.e. 



THE ACTION OF ACIDS ON CELLULOSE 



299 



ucts are usually stable at 125 to 130°. l Bay 2 states that the 
pure product is stable even at 180°. 

Action of Acids and Alkalis. — It has been claimed for hydro- 
cellulose that it is very resistant towards acids and alkalis. A 
product, highly resistant to these reagents, is obtained by treating 
cellulose with hydrochloric acid at a temperature of 60 to 70°, 
in the presence of a quantity of potassium chlorate insufficient 
to convert the cellulose to oxycellulose. 3 Boiling dilute acid 
had little effect on the hydrocelluloses prepared by Biittner and 
Neumann, 4 while boiling with caustic alkali produced only a 
yellow coloration. 

Hydrocelluloses prepared by various methods usually show 
a high solubility in alkali. Girard's hydrocellulose, prepared 
with hydrochloric acid gas, when boiled with 10 parts of 10 per 
cent sodium hydroxide, showed increased solubility with the 
length of heating; 85 per cent remained undissolved after 5 
minutes and 55 per cent after 30 minutes. 5 Schwalbe 6 found 
that Girard's hydrocellulose had the following solubilities in 
boiling 15 per cent sodium hydroxide: 



Hydrocellulose, 


NaOH solution, 


Time of boiling, 


Undissolved resi- 


grams 


grams 


minutes 


due, per cent 


10 


200 


10 


48 


10 


200 


20 


42 


10 


200 


30 


40 


10 


200 


40 


37 


10 


400 


60 


33 



Strong caustic soda dissolves about one-third of hydrocellu- 
lose, the residue being mercerized and its copper number greatly 
reduced. Of the dissolved portion, about 85 per cent can be 
reprecipitated with acids. 7 The hydrocelluloses prepared by 



1 H. Ost, Z. angew. Chem., 19 (1906), 994; Ann., 398 (1913), 313. 

2 C. Bay, Diss. Giessen (1913), p. 33. 

3 "Fabrik Chemischer Preparate von R. Sthamer," G. P. 123122 (1900). 

4 G. Buttner and J. Neumann, Z. angew. Chem., 21 (1908) 2609. 
6 C. G. Schwalbe, Z. angew., Chem., 20 (1907), 2171. 

6 C. G. Schwalbe, Z. angew. Chem., 22 (1909), 156. 

7 H. Jentgen, Z. angew. Chem., 23 (1910), 1545. 



300 CHEMISTRY OF CELLULOSE AND WOOD 

Qvist 1 were soluble to the extent of 41.4 to 66.5 per cent in 10 
per cent caustic soda. Bay 2 found solubilities of 54.94 to 58.80 
per cent, after heating for 2 hours with 10 per cent caustic soda. 
On the other hand, Knoevenagel's hydrocellulose is completely 
soluble in 8 per cent caustic soda in the cold. 

Hydrocellulose, on heating with lime water, gave the same or 
similar products as did oxycellulose, namely, isosaccharinic acid 
and an amorphous salt of what was apparently dihydroxybutyric 
acid. 3 Hydrocellulose, as usually prepared, behaves as a mixture 
of cellulose and hydrocellulose, the latter being removable by 
heating with solutions of the alkaline earths. 4 Girard's hydro- 
cellulose on boiling with lime showed a loss in weight of 11.6 
per cent in 30 minutes, and 24.4 per cent in 12 hours. 5 

Hydrocellulose, when heated with alkali under pressure, gave 
large yields of acetic acid. 6 With potassium carbonate solution 
at 100 to 110°, 14 per cent of acetic acid was obtained. Potas- 
sium hydroxide at 150° gave 19.5 per cent of acetic acid, while 
heating for 8 hours in a closed vessel gave 41.7 per cent. 

A hydrocellulose prepared by Vignon 7 gave 9.7 per cent of 
glucose when 2 grams were heated on the water bath for 6 
hours with 50 cubic centimeters of hydrochloric acid (sp. gr. 
1.125) and 200 cubic centimeters of water. Its heat of combus- 
tion was 4006 calories. Hydrocelluloses prepared by boiling 
cellulose with hydrochloric acid, with and without stannous 
chloride, were capable of giving the same amount of furfural, 
namely, 0.86 per cent. 8 

Schwalbe 9 found that Girard's hydrocellulose was more resist- 
ant to hydrolysis with 5 per cent sulphuric acid than cotton. 
The copper number due solely to hydrolysis (hydrolysis differ- 



1 W. Qvist, I.e. 

2 C. Bay, I.e. 

3 J. J. Murumow, J. Sack, and B. Tollens, Ber., 34 (1901), 1431. 

4 C. G. Schwalbe and E. Becker, J. prakt. Chem., 100 (1919), 19. 

5 H. Ost and R. Bretschneider, Z. angew. Chem., 34 (1921), 422. 

6 C. F. Cross and E. J. Bevan, J. Soc. Chem. Ind., 11 (1892), 966-969. 

7 L. Vignon, Compt. rend., 131 (1900), 708. 

8 L. Vignon, Compt. rend., 126 (1898), 1355. 

» C. G. Schwalbe, Z. angew. Chem., 22 (1909), 201. 



THE ACTION OF ACIDS ON CELLULOSE 301 

ence) was 0.9 for hydrocellulose. Bay 1 found hydrolysis differ- 
ences of 3.69 to 5.25. 

Certain forms of hydrocellulose appear to be very resistant to 
further hydrolysis with dilute acids under pressure or with con- 
centrated acids in the cold. This difference has been attributed 
to the formation of reversion products by the interaction of 
reducing sugars with the residue in the colloidal phase; hence 
the so-called hydrocellulose remaining after the prolonged heating 
of cellulose with dilute acids is not in the strict sense an inter- 
mediate degradation product. 2 Koerner 3 found that hydrocell- 
ulose gave a much higher yield of dextrose, as determined by 
fermentation to alcohol, than did sulphite cellulose when heated 
with dilute acid under pressure; the yields of alcohol were 17.95 
and 12.83 per cent respectively. Krull 4 subjected Girard's 
hydrocellulose to hydrolysis with strong hydrochloric acid and 
obtained only 13.2 per cent of sugar; the residue amounted to 
77.28 per cent. A preliminary treatment with chlorine increased 
the yield of sugar to 69.6 per cent. On the other hand, Knoeve- 
nagel's hydrocellulose gave over 95 per cent of fermentable 
glucose when hydrolyzed with 41 per cent hydrochloric acid. 5 

Solubility. — Hydrocellulose dissolves easily in all the known 
cellulose solvents. Girard 6 found that it dissolved instantly in 
Schweizer's reagent. Some confusion has originated on this 
point. The statement of Bronnert 7 that it dissolved only slightly 
in Wright's solution has been interpreted 8 as referring to Schwei- 
zer's reagent. In concentrated cuprammonium solution, hydro- 
cellulose forms 10 to 15 per cent solutions. 9 The cuprammonium 
solutions show remarkably low viscosities. 10 



1 C. Bay, I.e., p. 32. 

2 A. Wohl and K. Blumrich, Z. angew. Chem., 34 (1921), 17-18. 

3 T. Koerner, Z. angew. Chem., 21 (1908), 2357. 

4 H. Krull, Diss. Danzig (1916), p. 42. 

5 E. Heuser and F. Eisenring, Cellulosechemie, 4 (1923), 25. 

6 A. Girard, Ann. chim. phys., [5] 24 (1881), 364. 

7 E. Bronnert, G. P. 109996 (1899). 

8 C. G. Schwalbe, Z. angew. Chem., 20 (1907), 2171. 

9 H. Jentgen, Z. angew. Chem., 23 (1910), 1546. 
10 H. Ost, Z. angew. Chem., 24 (1911), 1896, 



302 CHEMISTRY OF CELLULOSE AND WOOD 

Action of Dyes. — Girard 1 noted that hydrocellulose had a 
greater affinity for dyes than ordinary cotton. According to 
Witz, 2 it is not dyed by basic dyestuffs. Schwalbe 3 found that 
ponceau 2R, brilliant crocein M, and patent blue, dyed hydro- 
cellulose, prepared by Girard's method with 3 per cent sulphuric 
acid, more strongly than cotton. With mikado orange there 
was no difference, while with benzopurpurine 10B, hydrocellulose 
was dyed more weakly than cotton. Hydrocellulose and its 
nitrate took up more methylene blue than cotton cellulose and 
its nitrate. 4 

Harrison 5 performed a series of experiments in which dilute 
sulphuric acid was allowed to dry on cotton, and found that the 
resulting products took up more methylene blue and less direct 
dyestuffs than cotton, properties usually attributed to oxycellu- 
lose. He concluded that the acid treatment probably formed 
colloidal hydrocellulose. 

The various discrepancies, with respect to the affinity of hydro- 
cellulose for dyestuffs, are without doubt due to the lack of uni- 
formity of the hydrocelluloses prepared by various authors. 
Not only the amount of actual hydrocellulose, but the presence 
of gelatinized cellulose, some oxycellulose, and the fixation of a 
portion of the acid, particularly sulphuric acid, used in converting 
the cellulose to hydrocellulose, must be considered. 

When cotton fabrics are subjected successively to printing 
with a strong solution of oxalic acid thickened with dextrin, air 
drying, and washing, the printed parts, while not tendered, show 
a decreased affinity for direct dyes and a marked affinity for 
methylene blue. 6 Boiling with dilute sodium hydroxide destroys 
this property. Cellulose appears to be capable of combining 
with small amounts of oxalic, malonic, and hexylmalonic acids, 
but not with tartaric, citric, succinic, and glutaric acids. It 

1 A. Girard, I.e., p. 369. 

2 G. Witz, Bull. Rouen, 11 (1883), 220. 

3 C. G. Schwalbe, "Chemie der Cellulose" (1911), p. 219. 

4 E. Berl and R. Klaye, Z. Schiess-u. Sprengstoff., 2 (1907), 381. 

5 W. Harrison, J. Soc. Dyers Colourists, 28 (1912), 238-239; cf. E. 
Grandmougin, Z. Farben-Ind., 3 (1907), 6. 

6 E. Knecht, Seventh Int. Cong. Appl. Chem. London, Sec. IVA (1909), 51. 



THE ACTION OF ACIDS ON CELLULOSE 303 

has been further shown 1 that when hydrocellulose is prepared by 
treating cellulose with dilute sulphuric acid (5 to 10 per cent) 
and drying at low temperatures (40 to 50°), the cellulose fixes 
0.97 to 1.14 per cent of sulphuric acid, which acts as a mordant 
for basic dyes. 

Hydrocellulose Nitrate. — Hydrocellulose nitrate contains as 
much nitrogen as cellulose nitrate, but it is more rapidly attacked 
by potassium hydroxide. It reduces Fehling's solution and 
shows slight aldehyde properties with Scruff's reagent. 2 The 
acetone solutions of hydrocellulose nitrate show a very low 
viscosity. 3 

The nitrogen content of a nitrated hydrocellulose was 13.25 
per cent in comparison with 13.5 per cent for the corresponding 
derivative from cotton. Hydrocelluloses prepared from sulphite 
cellulose and cotton, by the action of nitric acid of sp. gr. 1.1 
(17.1 per cent) in the cold for about 1 year, when nitrated in the 
cold with nitric acid of sp. gr. 1.4 to 1.5, gave xyloidins having 
a nitrogen content almost identical with that of the xyloidins 
prepared from cotton. 4 

Formula. — Girard gave hydrocellulose the formula (C 6 Hi O 5 )2.- 
H 2 0. Stern 5 erroneously denied the formation of hydrocellulose 
by the action of dilute acids. He obtained for the treated cellu- 
lose analyses agreeing closely with C 6 Hi O5, and concluded that 
Girard's hydrocellulose must have contained free acid and water. 
Biittner and Neumann 6 obtained for hydrocellulose dried at 
100° the general formula (C 6 Hio05)x.H 2 0, in which x is 2, 3, and 
6, depending on the method used for its preparation. Ost 7 
arrived at the formula (C 6 Hio05)6.H 2 for material dried at 
120°, but later, 8 after a careful series of analyses, came to the con- 

1 E. Knecht and F. P. Thompson, J. Soc. Dyers Colourists, 37 (1921), 
270-272. 

2 L. Vignon, Compt. rend., 126 (1898), 1658. 

3 E. Berl and R. Klaye, Z. Schiess-u. Sprengstoff., 2 (1907), 381. 

4 C. Haeussermann, Z. angew. Chem., 26 (1913), 456. 

5 A. L. Stern, /. Chem. Soc, 85 (1904), 336-340. 

6 G. Buttner and J. Neumann, Z. angew. Chem., 21 (1908), 2609; 22 
(1909), 585. 

7 H. Ost, Z. angew., Chem. 19 (1906), 994; H. Ost and F. Westhoff, 
Chem. Ztg., 33 (1909), 198. 

8 H. Ost, Ann., 398 (1913), 319. 



304 CHEMISTRY OF CELLULOSE AND WOOD 

elusion that hydrocellulose dried at 120° could not be distin- 
guished in elementary composition from ordinary cellulose. 

Nature of Hydrocellulose. — The nature and the composition 
of hydrocellulose have been the subject of considerable specula- 
tion. Jentgen 1 thought that hydrocellulose prepared with 
sulphuric acid of sp. gr. 1.453 to 1.530 was a mixture of hydro- 
cellulose and oxy cellulose together with unchanged cellulose. 

Hauser and Herzfeld 2 see in hydrocellulose an adsorption 
compound of variable composition, consisting of cellulose with 
its hydrolytic products, such as dextrins and dextrose. The 
property of brittleness is due to the presence of gelatinized cellu- 
lose, or its gelatinized degradation products, which are brittle in 
the dry state. Girard's hydrocellulose was extracted with water 
in a Soxhlet extractor and the cupric reducing power of the residue 
was determined at intervals of 12 hours. The original copper 
number of 5.4 decreased rapidly during the first hours of extrac- 
tion, dropping to 0.8 after 93 hours. When the residue was 
again treated by Girard's method, the copper number of 4.0 
decreased to 0.8 after 71 hours' additional extraction. Judging 
from the extraction curve, the hydrocellulose behaved as a gel 
containing adsorbed degradation products of cellulose. Glucose 
was identified in the aqueous extract. Harrison 3 had previously 
advanced the opinion that hydrocellulose consisted of the hydro- 
lytic products of cellulose adsorbed by peptized cellulose. 

By boiling hydrocellulose with Fehling's solution, a residue of 
pure cellulose is obtained after treatment with acid to remove the 
cuprous oxide, and washing with water. 4 The residue from the 
treatment of hydrocellulose with alkalis shows a considerable 
difference from ordinary cellulose. Ost and Bretschneider 5 
found that the viscosities of cuprammonium solutions of hydro- 
cellulose and the residue from boiling the latter with lime water 
were practically identical. The viscosity of a solution of a 
nitrating cotton that had been boiled with lime water decreased 

1 H. Jentgen, Z. angew. Chem., 23 (1910), 1545. 

2 0. Hauser and H. Herzfeld, Chem. Ztg., 39 (1915), 689-690; cf. 
C. G. Schwalbe and E. Becker, J. prakt. Chem., 100 (1919), 19. 

3 W. Harrison, J. Soc. Dyers Colourists, 28 (1912), 359. 

4 W. Netthofel, "Beitrage zur Kenntnis der Zellulose," Diss. Berlin 
(1914), p. 28. 

5 H. Ost and R. Bretschneider, Z. angew. Chem., 34 (1921), 422-423. 



THE ACTION OF ACIDS ON CELLULOSE 305 

after standing 7 days to the value for hydrocellulose. Acetyla- 
tion of the hydrocelluloses and residues, as well as of the nitrating 
cotton and residue, and the determination of the percentage of 
the acetylated product soluble in acetone, showed that the solu- 
bilities of the hydrocelluloses and their residues were practically 
identical. The conclusion is drawn that the residue is essentially 
the same as the original hydrocellulose, differing therefrom only 
in a lower copper number. 

According to Heuser, 1 true hydrocellulose consists of a dimeric 
anhydrocellobiose molecule in which an active carbonyl group 
is formed at the expense of an anhydride position, but without 
depolymerization of the dimeric molecule. The cellulose 
recovered from cellulose acetate has long been recognized as 
hydrocellulose. Barnett 2 has obtained what he considers 
definite hydrazones of the cellulose acetates with phenylhydra- 
zine and p-bromphenylhydrazine. 

Amyloid and Parchment. — It was observed by Schleiden 3 that 
when plant fibers were treated for half a minute with a mixture 
of three parts of concentrated sulphuric acid and one part of 
water, then iodine added, they were colored blue. The supposed 
transformation into starch was refuted by Liebig, 4 who showed 
that if the gelatinous cellulose obtained by Schleiden's method 
was washed after the acid treatment, tincture of iodine did not 
give a blue color. When the acid was filtered rapidly from the 
cellulose, and iodine solution added to the clear filtrate, a dark- 
blue precipitate was obtained. Water added to the filtrate 
precipitated a white flocculent substance that gave the blue 
color with iodine. 

The interesting discovery was made by Poumarede and Figuier 5 
that, when filter paper was dipped into sulphuric acid of 66° Be. 
(93.5 per cent) for half a minute and then into dilute ammonia to 
arrest the reaction, the paper resembled animal parchment. It 
was called "papyrine." Similar products were later known as 
vegetable parchment and amyloid. 

1 E. Heuser and W. von Neuenstein, Cellulosechemie, 3 (1922), 89. 

2 W. L. Barnett, /. Soc. Chem. Ind., 40 (1921), 61-63T. 

3 M. J. Schleiden, Ann., 42 (1842), 302. 

4 J. Liebig, Ann., 42 (1842), 306-309. 

5 J. A. Poumarede and L. Figuier, Compt. rend., 23 (1846), 918. 



306 CHEMISTRY OF CELLULOSE AND WOOD 

Strong sulphuric acid produces various products. Bechamp 1 
found that the dissolved cellulose could be precipitated with 
water and obtained in two water-soluble forms, "wood dextrin," 
and a gelatinous mass soluble in cold water. The latter product 
is very similar to the soluble cellulose of Guignet. 2 Hoffmann 3 
thought that the paper underwent no chemical change by being 
parchmentized. 

Paper can be parchmentized with nitric acid and various salts. 
Filter paper is remarkably toughened by moistening with nitric 
acid of sp. gr. 1.42 and subsequently washing with water. The 
strength is increased tenfold. The paper contains no nitrogen. 4 
Rankin 5 states that paper so treated is nitrated to some extent. 
During drying there is a linear shrinkage of about 10 per cent. 

Ferwer 6 treated 1 part of cotton with 30 parts of a mixture of 
4 parts of concentrated sulphuric acid and 1 part of water in a 
mortar; in half a minute he obtained a stiff jelly which in 15 
minutes gradually liquefied. Water precipitated a structureless, 
gelatinous mass, amyloid, which was colored blue by iodine, but 
the color disappeared on washing with water. The mass swelled 
in water to a paste which dried to thin, transparent films when 
spread upon glass. The acid solution gave dextrins and sugar 
on long standing. 

Guignet 7 obtained a cellulose forming a colloidal solution 
with water by treating filter paper or cotton with sulphuric acid 
of 50° Be. (62.5 per cent). The transparent, gelatinous mass 
did not change in the presence of an excess of acid at ordinary 
temperature but at about 100° dextrins were formed rapidly. 
The acid was removed by washing with water, finally with 
absolute alcohol. The washed product, dried at a low tempera- 
ture to remove the alcohol, gave a milky solution on digestion 
with water. The solution was difficult to filter, did not settle 
out on standing, and remained colloidal when heated. The 

1 A. Bechamp, Compt. rend., 42 (1856), 1213. 

2 C. E. Guignet, Compt. rend., 108 (1889), 1258-1259. 

3 A. W. Hoffmann, Ann., 112 (1859), 243. 

4 E. Francis, J. Chem. Soc., 47 (1885), 183-185; C. Beadle, Chem. News, 
112 (1915), 143-144. 

5 W. R. Rankin, Pharm. J., 95 (1915), 36. 

6 J. Ferwer, Dinglers polytech. J., 159 (1861), 218-221. 

7 C. E. Guignet, I.e. 



THE ACTION OF ACIDS ON CELLULOSE 307 

solution was slightly dextro-rotatory but did not reduce Fehling's 
solution. The semitransparent films, obtained by drying 
the solution on marble coated with vaseline, passed into colloidal 
solution again when treated with water. The films became insol- 
uble in water when plunged into sulphuric acid of 55 to 60° Be. 

Parchment paper consists of ordinary cellulose coated with 
colloidal cellulose (hydrocellulose). Guignet explained the 
disintegrating action of hot water on thin parchment paper to 
removal of colloidal cellulose. Thicker paper is not so affected, 
owing to the use in its manufacture of a stronger acid than that 
producing soluble cellulose. 

While studying the saccharification of cellulose, Flechsig 1 
obtained an amyloid by dissolving cotton in 69 per cent sulphuric 
acid. The cotton (250 grams) was treated with a cold mixture 
of 250 grams of sulphuric acid (75 per cent S0 3 ) and 84 grams of 
water. After standing an hour, addition of water gave a precipi- 
tate of amyloid which was held to be an intermediate product in 
the transformation of cellulose into dextrin. 

Zemplen 2 obtained only 15 to 20 per cent of octacetylcellobiose 
from Flechsig's amyloid, in comparison with 50 to 55 per cent 
obtained by Klein 3 from hydrocellulose and cellulose. The 
conclusion was drawn that amyloid and hydrocellulose are not 
identical. 

When wood is dissolved in strong sulphuric acid (78 to 93 
per cent) at 10 to 40°, it is converted into "acid cellulose" which 
can be thrown down by dilution with water. According to 
Ekstrom, 4 the product has a pronounced acid character but not 
that of an aldehyde. Complete conversion to glucose is obtained 
by heating with sulphuric acid at pressures slightly above atmos- 
pheric. Acid cellulose is insoluble in water, but forms water- 
soluble sodium salts. Schulz, 5 working with cotton and 78 per 
cent sulphuric acid, was unable to confirm these results. Acid 
cellulose had no pronounced acid character and contained 
aldehyde groups, as shown by the high copper number of 11.2. 

1 E. Flechsig, Z. physiol. Chem., 7 (1883), 523-540. 

2 G. Zemplen, Z. physiol Chem., 85 (1913), 180-192. 
3 F. Klein, Z. angew. Chem., 25 (1912), 1409. 

4 G. Ekstrom, G. P. 193112 (1907); 207354 (1909). 

5 W. Schulz, ''Zur Kenntnis der Cellulosearten," Berlin (1911), 100 pp. 



308 CHEMISTRY OF CELLULOSE AND WOOD 

Cotton fabric immersed in 62.5 per cent sulphuric acid at 0° 
for 10 minutes, and washed, forms a transparent leathery product. 1 
It takes up water very easily and is dyed much more intensely 
in an indigo bath than the ordinary fabric. 

A thorough investigation of the initial degradation products 
of cellulose by strong sulphuric acid was made by Schwalbe and 
Schulz. 2 Sulphuric acid of a concentration of 62.5 per cent 
gives Guignet's soluble cellulose, 69 per cent, " amyloid," and 
78 per cent, parchment, which is not identical with amyloid. 
Acids of higher concentration rapidly produce sulpholysis with 
the formation of acid esters. 

Guignet's Soluble Cellulose. — Five grams of air-dry cotton 
and 85 cubic centimeters of 62.5 per cent sulphuric acid are 
triturated in a mortar for 15 minutes. The mass is allowed to 
stand at room temperature with frequent stirring for 5 hours, 
170 cubic centimeters of water added, intimately mixed, and 
filtered on a cloth filter; the portion in colloidal solution in the 
filtrate is disregarded. The residue is washed free from acid 
by decantation, a procedure frequently requiring several days. 

Flechsig's Amyloid. — Five grams of absorbent cotton are 
gradually introduced into 40 grams of 69 per cent sulphuric acid 
during a period of 1 to 2 hours. The temperature, initially 6 to 
19°, is held below 30° by cooling. The mass becomes viscous, 
syrupy, then grayish white and opaque. Water precipitates a 
white flocculent mass, a large part of which passes into milky 
colloidal solution when suctioned on a cloth filter. The residue 
is washed free from acid by decantation. 

Parchment. — The cellulose is immersed in 78 per cent sulphuric 
acid for a few seconds, the time depending on the physical 
character of the cellulose. Filter paper is treated for 20 seconds 
and loose cotton for 10 seconds, then washed free from acid. The 
fibers become soft and transparent. The swelling of the fibers 
is usually followed by shrinkage. A portion of the cellulose 
passes into colloidal solution during the acid treatment, the solu- 
bility increasing with prolongation of the treatment. 



1 W. Minajeff, Z. Farben-Ind., 9 (1910), 65. 

2 C. G. Schwalbe and W. Schulz, Ber., 43 (1910), 913-917; Z. angew. 
Chem., 26 (1913) 499-501; W. Schulz, I.e. 



THE ACTION OF ACIDS ON CELLULOSE 



309 



Some of the properties of these modified celluloses are given 
in the following table : 



Material 


Copper 
number 
corrected 


Hydroly- 
sis number 
corrected 


Difference 

due to 
hydrolysis 


Water 

(dried 

at 105°), 

per cent 


Solubility 

in 10 per 

cent NaOH, 

per cent 




0.20 

1.95 

10.76 

18.8 
18.36 

7.11 
4.9 
11.2 


2.7 

4.45 

7.26 

20.5 
26.67 

17.63 

15.7 
30.4 


2.5 

2.5 

-3.5 

1.7 

8.3 

10.5 
10.8 
19.2 


5.2 
5.5 

2.4 

4.5 
12.0* 

9.3 
9.0 
7.0* 


8.6 




20.4 


Soluble Guignet cellulose. . . 
Parchmentized Guignet cell- 


70.1 
73.3 


Typical Flechsig amyloid . . . 
True parchment from: 

Absorbent cotton 

Filter paper 

Ekstrom acid cellulose 


99.4 

70.1 

18.3 

100.0 



* At 95°. 



Guignet's soluble cellulose and Flechsig's amyloid, like ordinary 
cotton, are not stained blue by iodine-potassium iodide solution, 
except in the presence of sulphuric acid; parchment paper, the 
closely related parchmentized Guignet cellulose, and Ekstrom's 
acid cellulose are all stained blue. With respect to colloidal 
properties, Guignet's cellulose stands first as a typical reversible 
hydrogel, followed by Flechsig's amyloid, and Ekstrom's acid 
cellulose. 

The maximum permissible drying temperature for Flechsig's 
amyloid and Ekstrom's acid cellulose is only 95°; the remaining 
modified celluloses may be dried at 105° without producing 
decomposition. 

As to solubility in alkali, Flechsig's amyloid and Ekstrom's 
acid cellulose are about equal, while decreasing solubility is 
shown by Guignet's cellulose, parchmentized Guignet cellulose, 
parchment paper, and cotton in their order. Flechsig's amyloid 
is most hygroscopic and Guignet's cellulose least so. 

With reference to hydrolysis differences, Guignet's cellulose 
with a negative difference (—3.5) is evidently a product inclined 
to reversion. Parchmentized Guignet cellulose is not, as Guignet 
assumes, a true parchment, as shown by the hydrolysis differ- 
ence, 1.7 in place of 10.7. On the other hand, Flechsig's amyloid 
with 8.3 approaches closely to true parchment, which supports 
the assumption that parchment is a mixture of cellulose, amyloid, 
and hydrocellulose. 



310 CHEMISTRY OF CELLULOSE AND WOOD 

Cellulose Dextrins and Sulphuric Esters. — In 1819, Bracon- 
not 1 found that when linen was left in contact with sulphuric 
acid (sp. gr. 1.827) at room temperature for some time, the solu- 
tion neutralized with lime, filtered, and the filtrate evaporated 
to dryness, there was obtained a gummy mass, which he called 
"acide-vegeto-sulfurique." The latter on hydrolysis was con- 
vertible into glucose. Blondeau de Carolles, 2 on studying the 
reaction, found that cellulose sulphuric esters of different compo- 
sition were obtainable according to the time in which the acid 
and the cellulose were in contact. The following water-soluble 
barium salts were isolated: 



Reaction period, hours 


Composition of the salts 


0.5 
12 
24 


C 18 H360 1 8(S0 3 ) 2 Ba0.2H 2 
C 10 H 2 oOio(S0 3 )2Ba0.2H 2 
C 4 H 8 4 (S03) 2 Ba0.2H 2 



It was established that the ratio of BaO to S0 3 was constantly 
1:2, a relationship confirmed by subsequent investigators. 
Fehling, 3 using linen, obtained a salt with the following 
composition : 

Per Cent 

BaO 5.0 

S0 3 5.3 

C 35.8 

H 5.9 

This corresponds with the formula C9 Hi 8 o0 9 o(S03)2.BaO. 
Marchand, 4 after leaving Swedish filter paper in contact with 
sulphuric acid for some weeks, isolated a "wood sulphuric acid" 
of the composition C 3 3H2808(S0 3 )2.CaO. It was observed by 
Bechamp 5 that the products obtained with concentrated 
sulphuric or hydrochloric acid resembled starch dextrin but had a 
lower rotatory power. He called one of the soluble products 
"wood dextrin," an unfortunate term that still persists. 

1 H. Braconnot, Ann. chim. phys., [2] 12 (1819), 172. 

2 Blondeau de Carolles, Ann., 52 (1844), 412. 

3 H. Fehling, Ann., 53 (1845), 134-136. 

4 Marchand, J. prakt. Chem., 35 (1845), 199-201. 
6 A. Bechamp, Ann., 100 (1856), 366-367. 



THE ACTION OF ACIDS ON CELLULOSE 



311 



Honig and Schubert 1 dissolved 10 grams of cotton in 40 cubic 
centimeters of concentrated sulphuric acid and studied the effect 
of time and temperature. Time had relatively little effect on 
the composition of the barium salts, but the optical activity rose 
decidedly with the temperature, as can be seen from the following 
data: 



Temperature 


Ba, 

per cent 


Rotation, 

[a]j 


+7° 


25.37 


- 3.65° 


10 


26.65 


- 1.62 


20 


26.75 


+37.05 


25 


24.34 


+44.10 


33 


25.40 


+57.36 


40 


25.29 


+72.99 



The composition of the salts and their optical activity were not 
affected unless the ratio of acid to cellulose fell below 2:1. The 
barium content increased slightly and the optical activity 
considerably with the time of reaction. A cellulose solution 
showed the following changes: 





Ba, per cent 


Mj 


After 0.5 hour 


24.34 
25.71 


+44.10° 


After 24 hours 


+67.77 







The barium salts were easily soluble in water and insoluble in 
alcohol; they reduced Fehling's solution and formed white to 
yellow, apparently crystalline powders. The free acids were 
very hygroscopic, amorphous substances which on boiling with 
alcohol decomposed into sulphuric acid and cellulose dextrins. 
General formulae are given for the acid esters and salts, but these 
must be accepted with reservation, since ultimate analyses were 
not made. 

Acid esters: CenHionOsn-sCSOOx. 

Salts: C te Hi „O6n-*(SO 4 M')*. 

When one part of cellulose and two parts of sulphuric acid 
were allowed to stand 30 minutes, diluted with ten parts of 

1 M. Honig and S. Schubert, Monatsh., 6 (1885), 708-728. 



312 CHEMISTRY OF CELLULOSE AND WOOD 

absolute alcohol and let stand 24 hours, acid esters were deposited; 
on boiling the latter with water, a series of compounds were 
obtained from amyloid to dextrins, according to the tempera- 
ture. 1 On increasing the reaction temperature from 3 to 40°, 
the optical rotation of the resulting products increased from [a]j = 
+6.4° to +127.72°; the solubility in water increased and the 
coloration with iodine passed from blue through bluish violet, 
violet, and red to colorless. The color reactions greatly resembled 
those obtainable with the degradation products of starch. 

Stern 2 dissolved cotton in sulphuric acid, and after a total 
reaction period of 7 hours, he prepared the barium salt of cellu- 
lose disulphuric acid, C 6 H 8 05(S0 3 )2Ba. The free acid obtained 
by removing the barium with sulphuric acid was very unstable 
and difficult to isolate. Hydrolysis of the barium salt with 2 
per cent sulphuric acid for 30 minutes gave glucose and a product 
corresponding with Ci 8 H 2 80i5.(S03H) 2 ; longer hydrolysis led 
to glucose and compounds with a diminishing content of sulphuric 
acid. The first product of hydrolysis was the ester, C 6 H 9 5 .- 
S0 3 H, which, like the parent substance, had no action on 
Fehling's solution. 

Cellulose dextrins were prepared by Yllner 3 by dissolving 
Swedish filter paper in 80 per cent sulphuric acid, and after 12 
hours, neutralizing with chalk. The glucose was removed by 
fermentation with yeast. The free dextrins were obtained by 
boiling the calcium salts with alcohol and fractionally precipi- 
tating them from aqueous solution with alcohol. Contrary 
to the findings of Honig and Schubert 4 the cupric reducing power 
of the dextrins increased with the increase in rotatory power. 
The highest rotation observed for any of the dextrins was 
[a] D = +107.5°. 

The influence of time, temperature, amount and concentra- 
tion of acid, on the formation of dextrin esters was studied by 
Muhlmeister. 5 The effect of time is shown in the table below. 

1 M. Honig and S. Schubert, Monatsh., 7 (1886), 455-468. 

2 A. L. Stern, J. Chem. Soc, 67 (1895), 74-90. 

3 C. A. Yllner, Z. angew. Chem., 25 (1912), 103-107. 

4 M. Honig and S. Schubert, I.e., 463. 

5 H. Muhlmeister, " Beitrage zur Hydrolyse und Sulfolyse der Zellulose," 
Diss. Hannover (1913), 44 pp. 



THE ACTION OF ACIDS ON CELLULOSE 



313 



The cellulose was dissolved in 70 per cent sulphuric acid at 15° 
and the esters precipitated with alcohol-ether mixture. With 



Influence of Time on the Properties of Dextrin ] 


]STERS 










Milligrams 




Hours 


Yield 

from 1.887 

grams 

cellulose 


Corrected 
copper 
number 


Sulphuric 

acid, 
per cent 


reduced 
copper from 

25 cubic 
centimeters 
of solution 


Solubility 
in water 


1 


1.847 


12.4 


6.6 




_ 


2 


1.761 


19.7 


8.1 




— . 


3 


1.651 


24.6 


9.6 


1.7 


+ 


4 


1.566 


22.4 


11.7 


3.6 


+ 


5 


1.410 


25.9 


11.4 


5.6 


+ 


6 


1.274 


25.9 


12.8 


6.8 


+ 


7 


1.477 


25.2 


10.5 


3.9 


+ 


8 


1.322 


27.8 


11.0 


5.7 


+ 


9 


1.074 


27.5 


11.0 


7.7 


+ 


10 








9.2 


+ 


11 


0.534 


34.3 


12.1 




+ 


12 








11.4 


+ 



time the yield of esters decreased, while the sulphuric acid 
content and reducing power of the esters and dextrins remaining 
in solution in the alcohol-ether mixture, increased. When 75 
cubic centimeters of 70 per cent sulphuric acid were allowed to 
act on 2 grams of cotton for 6 hours, the solution diluted with 
an equal volume of water and let stand 24 hours, the esters, on 
precipitation, had a sulphuric acid content of 26.2 per cent, which 
corresponds with C12H19O10.SO3H. 

Thirty grams of cotton were allowed to react with 300 cubic 
centimeters of 70 per cent sulphuric acid for 20 hours. Addition 
of 2 liters of alcohol precipitated 3.1 grams of an ester having a 
copper number of 37.8 and a rotation in aqueous solution of 
Mz>°° = +15.4°. The ester, though originally soluble in water, 
was rendered partially insoluble by boiling with alcohol. There 
was obtained 40 per cent of water-soluble dextrin and 60 per cent 
of water-insoluble dextrin; the copper numbers were 57.9 and 



314 



CHEMISTRY OF CELLULOSE AND WOOD 



43.9 respectively. Ether precipitated from the alcoholic solu- 
tion two esters having the following properties: 





Copper number 


r i20 o 


H 2 S04 content 


1 

2 


33.4 
118.9 


+21.1° 
+54.3 


18.9 



The cellulose dextrins, on acetylation, gave products containing 
63 to 72 per cent of acetic acid. The acetates soluble in ether 
corresponded with glucose pentacetate and glucose tetracetate. 
Cellobiose octacetate could not be detected. 

Ost 1 is of the opinion that Honig and Schubert did not obtain 
true dextrins but rather condensation products resulting from 
reversion of the dextrins by the concentrated sulphuric acid. 
Concentrated sulphuric acid forms dextrin esters, but for the 
sulpholysis of cellulose, 70 per cent sulphuric acid is a better 
reagent. Acid of a concentration of 56 per cent does not form 
esters with cellulose in the cold. 2 

Esparto cellulose, containing about 20 per cent of pentosans, 
gave on sulpholysis esters in which Ba was approximately 
equal to S0 4 , the ratio being Ba 2 :(S0 4 )3. This indicates that 
some of the barium must be combined with hydroxyl groups of a 
sufficiently acid character to decompose barium carbonate. 3 
The stable character of the sulphate residues is indicated by the 
fact that their elimination was complete only after boiling with 
concentrated hydrochloric acid, in the presence of barium 
chloride, for some hours. On treating 60 grams of esparto 
cellulose with 300 cubic centimeters of 73 per cent sulphuric acid 
for 24 hours, diluting with water, and fractionally precipitating 
with alcohol, two fractions were obtained as follows: 



Number 


Yield, 
grams 


Ba, 
per cent 


S0 4 , 
per cent 


Furfural, 
per cent 


[ah 


A 
B 


37 

8 


8.6 
8.6 


7.9 
10.6 


8.3 
8.0 


+47.8 to 45. 5° 
+55.3 to 51. 6° 



1 H. Ost, Ann., 398 (1913), 322. 

2 E. Flechsig, Z. physiol. Chem., 7 (1883), 528. 

3 M. Cunningham, J. Chem. Soc, 113 (1918), 173-181. 



THE ACTION OF ACIDS ON CELLULOSE 



315 



Cotton cellulose was hydrolyzed with 40 per cent hydrochloric 
acid for 48 hours. The followine; lead salts were isolated: 



Number 


Pb, per cent 


CI, per cent 


[ah 


A 
B 


8.3 
9.4 


6.2 

9.8 


+42.5 to34.5° 
+26.7 to 17.3° 



It is interesting to note that the above salts contain chlorine, 
and that esparto cellulose when hydrolyzed under similar condi- 
tions gave very similar esters containing chlorine. 

An extensive investigation of cellulose dextrins has been made 
by Samec and Matula. 1 The raw material was sulphite cellu- 
lose. The authors used sulphuric acid (50 to 85 per cent) and 
phosphoric acid (76 to 85 per cent) at temperatures of 50 to 
70° for variable periods up to 6 days. The course of the reaction 
was followed by: (1) the reaction with iodine, the color changing 
from blue through violet to brown; (2) the change in the propor- 
tion of the products soluble in water; (3) the effect of alcohol in 
which 

a. The dextrin esters gelatinize and the dextrin precipitates. 

b. The dextrin esters and dextrins are insoluble. 

c. The dextrin esters are soluble but the dextrins insoluble. 

d. Both dextrin esters and dextrins are soluble. 
Five different dextrins were prepared as follows: 

A. By the action of 50 parts of 55 per cent sulphuric acid on 2 
parts of cellulose at 20° for 45 minutes. 

B. In the same way as A, except that the reaction was prolonged 
until water caused no separation, i.e., 4 to 5 hours. 

C. By the action of 5 parts of 55 per cent sulphuric acid on 2 
parts of cellulose for 24 hours. 

D. By the action of 20 parts of 85 per cent phosphoric acid on 
1 part of cellulose for 1 hour at 50°. 

E. By the action of 25 parts of 80 per cent phosphoric acid 
on 1 part of cellulose for 24 hours at 50°. 

The reaction products were dialyzed until free from electrolytes, 
a process requiring weeks. Some of the properties of these 
dextrins are given below. Judging from titration with sodium 

1 M. Samec and J. Matula, Kolloidchem. Beihefte, 11 (1919), 37-73. 



316 



CHEMISTRY OF CELLULOSE AND WOOD 



hydroxide, one — S0 4 H group is combined with 41 to 42 C 6 Hi O 5 
groups in dextrin A, and with six C 6 H 10 O5 groups in dextrin C. 





Solubility 


Iodine 
reaction * 


Cop- 
per 

num- 
ber 


Dialyz- 

able 
portion, 
per cent 


Molecular 
weight of 


Dextrin 


In 

water 


In 

alcohol 


non-dialyz- 

able 

residue 


A 
B 
C 
D 
E 


Slight 

Soluble 

Soluble 

Slight 

Soluble 


Gelatinous 

Insoluble 

Insoluble 

Gelatinous 

Soluble 


Blue 

Bluish violet 
Bluish violet 
Blue 
Dark violet 


15.8 
36.9 
49.8 

8.6 


7 
75 
50 

70 


6800 

2400 

6400 

15000 

3000-4000 



* With the sulphuric acid solutions. 

The action of diastase on the dextrins was also studied. 
Dextrin A in the form of a suspension was not noticeably 
attacked; after 2 hours' action on the colloidal solution, the 
copper number was reduced from 15.8 to 6.9, at which point the 
dextrin became flocculent. Dextrin B resembled starch in that 
the viscosity decreased, the reducing power increased, and a 
fermentable sugar was formed after 2 hours. Dextrin C 
resembled B, while D was not attacked. Dextrin E differed 
only quantitatively from the dextrins prepared with sulphuric 
acid. 

When benzene or its homologs is added to a solution of cellu- 
lose in strong sulphuric acid, a reaction takes place to give an 
amorphous, infusible compound, insoluble in the ordinary 
solvents; these are called desoxyns. 1 The phenyl group is 
evidently directly connected with a carbon atom in the carbo- 
hydrate, since phenyldesoxyn, CeHyC^.CCeHs^, gives toluene on 
dry distillation, and benzoic acid on oxidation with potassium 
permanganate. 

Cellulose Acetates. — Cellulose acetate was first prepared by 
heating cellulose with acetic anhydride at a high temperature. 2 
Cross and Bevan 3 obtained the acetate by treating mercerized 
cellulose, or cellulose regenerated from its thiocarbonate, with 

•A. Nastjukoff, J. Russ. Phys.-Chem. Soc, 34 (1902), 231, 505; Z. 
Farben-Ind., 6 (1907), 70. 

2 M. P. Schutzenberger and L. Naudin, Compt. rend., 68 (1869), 814. 

3 C. F. Cross and E. J. Bevan, G. P. 85329 (1894) and 86368 (1895). 



THE ACTION OF ACIDS ON CELLULOSE 



317 



acetyl chloride below 30° and a catalyzer such as zinc acetate 
or magnesium acetate. The cellulose acetate was soluble in 
chloroform and gave transparent elastic films. Their view 1 
that the product was a tetracetate was based on a faulty method 
of analysis. 2 Unless there is hydrolysis of the cellulose, no 
derivative higher than the triacetate is obtainable. 

Lederer 3 prepared cellulose acetate by acetylating Girard's 
hydrocellulose ; also by first treating the cellulose with acetic 
acid and a little sulphuric acid at 60 to 70°, cooling, adding acetic 
anhydride, and allowing the reaction to proceed at room tem- 
perature. 4 Bayer and Company 5 employed a mixture of acetic 
acid, acetic anhydride, and sulphuric acid at a temperature 
below 50°. The acetates obtained by the two latter methods 
are identical and considered by Ost 6 as the triacetate of a hydro- 
cellulose, (CeHioC^e-H^O or C 3 6H 62 03i. The compound con- 
tained only 18 acetyl groups and not 19 or 20 as would be required 
if one molecule of water had combined through hydrolysis. 

It is impossible to arrive at a definite decision regarding the 
presence of a molecule of water in a cellulose aggregate of the 
dimensions of C 3 e by ultimate analysis; nor is there a method for 
determining acetyl groups which is sufficiently reliable for this 
same purpose. 





Acetic acid for 


Formula 


Diacetate, 
per cent 


Triacetate, 
per cent 


Tetracetate, 
per cent 


(C 6 H 10 O5)6 

(C 6 Hio0 6 )6-H 2 


48.76 
48.17 


62.47 
61.83 


72.70 

72 05 







Schwalbe 7 came to the conclusion that the cellulose acetates 
of Bayer and Company and of Lederer were identical, and were 
derivatives of hydrocellulose, owing to their reducing properties. 

1 C. F. Cross and E. J. Bevan, "Researches" (1895-1900), p. 80; cf. 
L. Vignon and F. G'erin, Compl. rend., 131 (1900), 590. 

2 H. Ost, Z. angew. Chem., 19 (1906), 998. 

3 P. Lederer, G. P. 118538 (1899) and 120173 (1900). 

4 P. Lederer, G. P. 163316 (1901). 

5 F. Bayer and Company, G. P. 159524 (1901). 

6 H. Ost, I.e. 

7 C. G. Schwalbe, Z. angew. Chem., 23 (1910), 433-441; 24 (1911), 1256- 
1260, 



318 CHEMISTRY OF CELLULOSE AND WOOD 

The copper numbers of the cellulose acetates were 5.1 to 6.3 
and 6.1 to 6.6 respectively. Acetylation by the Bayer method is 
preceded by the formation of hydrocellulose. If the acetylation 
is interrupted after 5 to 10 minutes, the acetylated portion has a 
reducing action on Fehling's solution. The later stages of acetyl- 
ation are without increase in reducing power. The products 
first formed by the Lederer process do not reduce Fehling's 
solution but do so towards the end of the reaction. Acetylation 
of the hydrocellulose proceeds at a far more rapid rate than the 
formation of hydrocellulose. 

Hydrocellulose was acetylated readily by heating with acetic 
anhydride and acetic acid, without the presence of sulphuric 
acid, while with sulphuric acid mercerized cotton was acetylated 
more slowly than normal cotton. 1 Hydrocellulose and an alkali- 
cellulose, prepared by heating cellulose 48 to 72 hours with 25 
per cent sodium hydroxide solution at 110 to 120° and washing, 
were esterified far more rapidly than normal cellulose. 2 When 
5 grams of material were acetylated by heating at 70° with 25 
grams of glacial acetic acid, 25 grams of acetic anhydride, and 
2.5 grams of anhydrous zinc chloride, hydrocellulose and the 
alkali cellulose required only 2 hours for complete acetylation 
while normal cellulose required 20 hours. When 0.5 gram of 
sulphuric acid was used as the catalyzer and the acetylation 
conducted at room temperature, the acetates from cellulose and 
hydrocellulose appeared to be identical after 24 hours. Increas- 
ing the sulphuric acid to 1 gram, gave acetylated products almost 
entirely soluble in alcohol, cellobiose octacetate predominating. 

Cellulose acetates insoluble in acetone are readily converted 
into soluble modifications by heating at temperatures above 
100°, in the absence of water, with ethylacetate, benzene, ace- 
tone, alcohol, acetylene tetrachloride, and similar solvents. 3 
Neutral salts may be used as catalyzers. In the complete 
absence of water the conversion reaction is greatly prolonged. 
By the use of bisulphates in the presence of water, the acetyl 
value of the esters is not lowered, nor is the copper number 
higher than for the insoluble form. Solutions of acetone-soluble 

1 H. Ost and F. Westhoff, Chem. Ztg., 23 (1909), 198. 

2 H. Ost and T. Katayama, Z. angew., Chem., 25 (1912), 1467-1470. 

3 E. Knoevenagel, Z. angew. Chem., 27 (1914), 505-509. 



THE ACTION OF ACIDS ON CELLULOSE 319 

cellulose acetates, in the presence of small quantities of catalysts, 
such as sulphuric acid and sulphoacetic acid, gel on standing. 1 
The cellulose acetate is then insoluble in acetone, acetylene 
tetrachloride, and lactic acid esters. The reversal of the solu- 
bility of cellulose acetates is accompanied by only minor changes 
in the copper number, indicating that hydrolysis does not take 
place to any appreciable extent. 

Between the cellulose acetates soluble in chloroform and those 
soluble in acetone, which are presumably hydrocellulose acetates, 
there is a whole series of intermediates. A slight physical or 
chemical change in the cellulose acetate produces very marked 
differences in the solubility of the ester and in the viscosity of 
its solutions. 

Cellulose triacetate, which is soluble in chloroform and tetra- 
chlorethane, shows an optical rotation of [a] D = ca.— 20° in 
these solvents, and [a] D = ca. — 50° in pyridine. Cellulose ace- 
tate deposits from its solution in tetrachlorethane large crys- 
talline needles containing some of the solvent with which it 
apparently enters into molecular union. 2 

Nascent sulphuryl chloride has been found to be a very active 
catalyst for the acetylation of cellulose and to produce a minimum 
of change in the cellulose acetate. 3 Immerse 5 grams of filter 
paper in 20 cubic centimeters of glacial acetic acid and 20 cubic 
centimeters of acetic anhydride containing 0.32 gram of chlorine; 
then add 2 cubic centimeters of acetic anhydride containing 0.26 
gram of sulphur dioxide, and stir. The mixture gelatinizes 
rapidly and solution is complete in 5 minutes. After an addi- 
tional 5 minutes' stirring, add 20 cubic centimeters of chloroform, 
pour into an excess of water, and heat to expel the chloroform. 
On breaking up the resulting globules by stirring, the cellulose 
triacetate deposits as a fine powder, which is easily washed free 
from acid. The ester contains 62.4 per cent of acetic acid. 

Wood cellulose, obtained by extracting wood with 40 per cent 
hydrochloric acid in the cold for 42 hours and precipitating with 
water, gives a cellulose triacetate producing brittle films. 4 Zinc 

1 E. Knoevenagel and K. Konig, Cellulosechemie, 3 (1922), 113-122. 

2 K. Hess, Z. angew. Chem., 37 (1924), 997. 

3 W. L. Barnett, J. Soc. Chem. Ind., 40 (1921), 8-10T. 

4 E. Hagglund et al., Cellulosechemie, 3 (1922), 13-19. 



320 CHEMISTRY OF CELLULOSE AND WOOD 

chloride and sodium ethylsulphate may be used as catalyzers, 
but sulphuric acid is unsatisfactory. Sulphite pulp under appro- 
priate conditions gives an ester suitable for films. With sulphuric 
acid, there is a great tendency to form products with a high 
copper number and approaching a tetracetyl derivative. An 
acetate soluble in acetone but insoluble in chloroform is obtained 
as follows : Treat 5 grams of sulphite pulp with a solution of 20 
grams of 100 per cent acetic acid, 0.5 gram of water, and 0.5 
gram of sodium bisulphate. Heat for 17 hours at 50 to 70°, 
cool, and add with stirring 25 grams of acetic anhydride. The 
temperature must not exceed 60°. After warming the viscous 
solution at 70° for about half an hour, the reaction is ended. 
Cool the mixture somewhat, add 5 to 6 cubic centimeters of 
water, and let stand at 50° for 65 to 70 hours; then precipitate 
the acetate by pouring the solution into an excess of water. 

Nauck 1 considers that for the preparation of useful acetates 
from wood pulp the physical properties of the pulp are more 
important than the chemical. The pulp should be in a finely 
divided condition. Acetylation proceeds the more uniformly 
the greater the specific surface. In this respect cotton behaves 
better than wood pulp, and dissolves more homogeneously 
during acetylation. 

Cellulose Nitrates. — The alleged monoesters of cellulose are 
shown by X-ray examination to be mixtures of esters and 
unchanged cellulose, products having the compostion of a mono- 
ester being accidental. 2 Acetylation and benzoylation of cellulose 
are surface reactions proceeding from the exterior to the interior 
of the fiber as the esters formed are removed. The speed of 
esterification depends more on the previous chemical and physical 
treatments to which the fiber has been subjected than on the 
degree of symmetry in the arrangement of the crystallites. 
Nitration and denitration are topochemical reactions, that is, 
take place as if in a homogeneous sytem. The denitrated fiber 
gives the same point diagram as the original fiber, showing that 
there has been no change in the position of the crystallites. 

Cellulose treated with a mixed acid consisting of 2 parts of 
concentrated sulphuric acid and 1 part of concentrated nitric 

1 W. Nauck, Cellulosechemie, 2 (1921), 61-63. 

2 R. O. Herzog and G. Londberg, Ber., 57 (1924), 329-332. 



THE ACTION OF ACIDS ON CELLULOSE 321 

acid is nitrated in 1 hour at a temperature of 15 to 20°. In 
mixed acid, the highest nitration is obtained when the system is 
represented by HN0 3 + n(H 2 S0 4 .H 2 0) and the lowest when 
nitric acid is present in the form of the hydrate, HNO3.H2O. 1 
The sulphuric acid, accordingly, performs the function of freeing 
the nitric acid from its hydrate. It was thought by Cross and 
Be van 2 that sulphuric acid esters of cellulose were formed, and 
that these were decomposed on washing with the formation of 
nitric esters. This view has been shown to be incorrect. 3 In 
mixed acid, the characteristic action of each acid is retarded. 
Cellulose dissolves rapidly in concentrated sulphuric acid, solu- 
tion being followed by hydrolysis. Nitric acid (91 to 95 per 
cent) rapidly forms insoluble nitrates containing 9.6 to 13.5 
per cent of nitrogen, hydrolysis and oxidation following on pro- 
longed standing. 

Vieille, 4 using nitric acid of sp. gr. 1.430 to 1.502, found that 
the more concentrated the acid, the greater was the esterification 
of the cellulose. It was, accordingly, assumed by Zachias 5 and 
Justin-Mueller 6 that cellulose nitrates were adsorption com- 
pounds. The nitric acid was dehydrated by the sulphuric acid 
to nitrogen pentoxide which was adsorbed by the cellulose. 
This is not true since the nitration of cellulose does not follow 
an adsorption curve. 7 With increasing concentration of nitric 
acid, the nitration advances step by step and not gradually. 
Furthermore, the reaction is not reversible, no nitric acid being 
removable from the nitrated cellulose with water. 

Knecht 8 obtained a labile cellulose nitrate which was apparently 
an adsorption compound, by treating cellulose with nitric acid 
of sp. gr. 1.415. Water decomposed the dried product with the 
formation of nitric acid and gelatinized cellulose. A nitric acid 



1 A. Saposchnikoff, Z. Schiess-u. Sprengstoff., 1 (1906), 453-456; K. 
Kullgren, Ibid., 3 (1908), 146-149. 

2 C. F. Cross and E. J. Bevan, Ber., 34 (1901), 2496. 

3 C. N. Hake and M. Bell, J. Soc. Chem. Ind., 28 (1909), 457-464. 

4 Vieille, Compt. rend., 95 (1882), 132-135. 

5 P. D. Zachias, Z. Farben-und Textilchem., 2 (1903), 233-239. 
6 E. Justin-Mueller, Kolloid-Z., 2 (1907), 49-51. 

7 A. Muller, Kolloid-Z., 2 (1907), 173-175. 
s E. Knecht, Ber., 37 (1904), 549-552. 



322 



CHEMISTRY OF CELLULOSE AND WOOD 



of higher specific gravity than 1.415 gave true esters. An addi- 
tion compound is first formed which in an excess of acid passes 
into the ester by the elimination of water. 1 

In the case of cellulose nitrates, the introduction of a nitric 
acid group results in an increment of 26.8 cubic centimeters in 
the molecular solution volume, while with polyatomic alcohols, 
such as glycerine, there is a decrease in the volume for each 
nitric acid group from the first to the third. The difference is 
explained by assuming a cyclic structure for cellulose. 2 

The various possible nitrates of cellulose, using a (CeHioOs^ 
molecule, are as follows: 



Cellulose 


Formula 


N, per cent 


Dodecanitrate 


C24H 2 802o(N02) 12 

C 2 4H29O20(NO 2 )ll 

C24H3o02o(N0 2 ) 10 

C24H3l0 2 o(N0 2 )9 

C24H 3 2O20(NO 2 )8 

C 24 H330 2 o(N0 2 )7 

C24H340 2 o(N0 2 )6 

C 2 4H350 2 o(N02)5 

C 2 4H360 2 o(N0 2 ) 4 


14 16 


Endecanitrate 


13.50 


Decanitrate 


12.78 


Enneanitrate 


11 98 


Octonitrate 


11.13 


Heptanitrate 


10.19 


Hexanitrate 


9.17 


Pentanitrate 


8.04 


Tetranitrate 


6.77 







A cellulose trinitrate should theoretically contain 14.16 per 
cent of nitrogen. Hoitsema 3 obtained a nitrate with 13.9 per 
cent of nitrogen by the use of pure, crystallized, anhydrous 
nitric acid. Lunge 4 obtained a cellulose nitrate containing 13.92 
per cent of nitrogen with a mixed acid of the composition: 
H 2 S0 4 , 63.35 per cent; HN0 3 , 25.31 per cent; H 2 0, 11.34 per 
cent. After a few months, the nitrogen content dropped to 13.50 
per cent, where it remained constant. The influence of water 
on the degree of nitration will be seen in the following table : 



1 C. Haeussermann, Z. Schiess-u. Sprengstoff., 3 (1908), 121-122; 
E. Knecht and A. Lipschitz, J. Soc. Chem. Ind., 33 (1914), 116-122. 

2 C. F. Cross and E. J. Bevan, Ber., 42 (1909), 2198-2204. 

3 C. Hoitsema, Z. angew. Chem., 11 (1898), 173-174. 

4 G. Lunge, J. Am. Chem. Soc, 23 (1901), 527-579. 



THE ACTION OF ACIDS ON CELLULOSE 
Influence of Water on Nitration 



323 



Number 


N, 
per cent 


Soluble 
in ether- 
alcohol, 
per cent 


Yield 

on 
cotton, 
per cent 


Acid mixture, per cent 


H 2 S0 4 


HNO s 


H 2 


1 


13.65 


1.50 


177.5 


45.31 


49.07 


5.62 


2 


13.21 


5.40 


176.2 


42.61 


46.01 


11.38 


3 


12.76 


22.00 




41.03 


44.45 


14.52 


4 


12.58 


60.00 


167.0 


40.66 


43.85 


15.49 


5 


12.31 


99.14 


159.0 


40.14 


43.25 


16.61 


6 


12.05 


99.84 


153.0 


39.45 


42.73 


17.82 


7 


11.59 


100.00 


156.5 


38.95 


42.15 


18.90 


8 


10.93 


99.82 


144.2 


38.43 


41.31 


20.26 


9 


9.76 


74.22 


146.0 


37.20 


40.30 


22.50 


10 


9.31 


1.15 


138.9 


36.72 


39.78 


23.50 


11 


8.40 


0.61 


131.2 


35.87 


38.83 


25.30 


12 


6.50 


1.73 




34.41 


37.17 


28.42 



The solubility of cellulose nitrates shows considerable variation. 
Solubility of Cellulose Nitrates* 





Solubility in 


N, per cent 


Acetone, 
ethylacetate, 


Ether-alcohol 


Absolute alcohol, 




amylacetate, 


2: 1, per cent 


per cent 




per cent 






13.1-13.4 


95-100 


Insoluble 


Insoluble 


12.75-13.1 


95-100 


<30 


Insoluble 


12.5-12.75 


95-100 


50-100 


<10 


12.0-12.5 


95-100 


95-100 


<50 


11-12 


95-100 


90-100 


50-100 


10-11 


95-100 


80-100 


<50 


9-10 


30-90 


30-900 


Insoluble 


7-9 


<30 


<30 


Insoluble 


3-7 


Insoluble 


Insoluble 


Insoluble 



* Ullmann, "Enzyk," 5 (1917), 96. 

Nitrates prepared from cotton and viscose showed practically 
identical nitrogen contents and solubilities in ether-alcohol 
mixture. 1 

1 A. Lusk and C. F. Cross, J. Soc. Chem. Ind., 19 (1900), 642-644. 



324 CHEMISTRY OF CELLULOSE AND WOOD 

In the presence of strong nitric acid the nitrogen content 
of the cellulose increases to a maximum and then decreases. 
This denitrification results in the formation of oxy cellulose. 1 
Vignon 2 obtained identical cupric reducing values for oxycellu- 
lose nitrates and cellulose nitrates. 

Strongly bleached cotton gave nitrates with 9.80 to 10.37 per 
cent of nitrogen in comparison with 11.8 to 11.92 per cent of 
nitrogen for cotton of high quality. 3 The oxy cellulose esters 
obtained by nitrating strongly bleached cotton produce a 
decrease in viscosity of normal esters. The addition of definite 
amounts of such solutions of low viscosity to a cellulose nitrate 
solution of high viscosity caused a reduction in viscosity out 
of all proportion to the addition. 4 

Solutions of cellulose nitrates of low viscosity may be obtained 
by using mercerized cellulose, hydrocellulose, and cellulose that 
has been modified by heating. 5 Absorbent cotton was heated 
at 100° for 60 hours in atmospheres of oxygen, hydrogen, and 
carbon dioxide. The outflow times of acetone solutions of 
nitrates prepared from the celluloses so heated were 463, 1063, 
and 1110 seconds respectively, while a nitrate solution prepared 
from the unheated cotton gave a reading of 23330 seconds. 

Cellulose treated with nitric acid of sp. gr. 1.473 to 1.50 
passes into solution with the formation of xyloidins. 6 Prepara- 
tion : Ten grams of cotton are saturated with water, pressed to a 
weight of 20 grams, and introduced with good cooling and 
stirring with a thermometer into 300 cubic centimeters of nitric 
acid of sp. gr. 1.50. The temperature rises to 25 to 30° but sinks 
rapidly with good cooling. The cellulose dissolves very rapidly 
and uniformly, but not completely. There is obtained about 315 
cubic centimeters of a yellow, honey-like liquid. The initially 
high viscosity decreases rapidly. Water precipitates snow- 
white films containing 8.8 per cent of nitrogen. 7 

X E. Knecht and A. Lipschitz, J. Soc. Chem. Ind., 33 (1914), 117. 
2 L. Vignon, Compt. rend., 131 (1900), 509-511. 

3 R. Namias, Mon. sci., 8 (1918), 5-6; cf. C. Piest, Z. angew. Chem., 21 
(1908), 2497-2499. 

4 C. Piest, Z. angew. Chem., 26 (1913), 24-30. 

6 E. Berl, Z. Schiess-u. Sprengstoff., 4 (1909), 81-83. 

6 C. Hauessermann, Z. Schiess-u. Sprengstoff., 1 (1906), 39; 3 (1908), 305. 

7 H. Jentgen, Z. angew. Chem., 25 (1912), 945. 



THE ACTION OF ACIDS ON CELLULOSE 325 

In nitric acid of sp. gr. 1.450 to 1.463, cotton fibers swell 
strongly, disintegrate, and pass into solution. Water produces 
almost complete precipitation, giving a product containing 6.5 
to 8.0 per cent of nitrogen. 1 Acid of sp. gr. 1.469 to 1.476 gives 
a viscous fluid, xyloidin, containing 8 to 9 per cent of nitrogen, 
which is insoluble in acetone and ether-alcohol mixture. With 
acid of sp. gr. 1.48 to 1.50, cotton without noticeable structural 
change gives pyroxylin containing 9 to 12.5 per cent of nitrogen. 
Sulphite pulp behaves similarly, giving with acid of sp. gr. 1.495 
a product with 11.1 per cent of nitrogen. 2 Acids of higher 
density have a strong parchmentizing action. Xyloidins, cellu- 
lose nitrates, and cellulose acetates, on standing in concentrated 
nitric acid for 3 to 4 months, are completely decomposed into 
water-soluble products. 3 Acetylcellulose on standing in fuming 
nitric acid for 14 days gave, presumably, a hydrocellulose nitrate 
containing 11.5 per cent of nitrogen. 4 

Cellulose nitrate dissolves rapidly on shaking with cupram- 
monium solution. 5 

Ammonium sulphide is one of the most efficient denitrating 
agents for cellulose nitrates. 6 

Collodion, treated with 10 per cent sodium hydroxide in the 
cold for 2 hours, is decomposed with the formation of oxypyruvic 
acid, CH 2 OH.CO.COOH, which yields a yellow osazone. 7 
Phenylhydrazine reacts with acetylcellulose nitrate to give the 
compound C24H 28 07(N0 3 )4.(CH3.COO) 8 .HN 2 .C 6 H5. 8 Alcoholic 
sodium hydroxide acting on cellulose nitrate gave 7 to 8 per cent 
oxypyruvic acid, and cellulose tetranitrate, the latter reacting 
with phenylhydrazine to give C 2 4H360i5.(N03)4.NH2.C 6 H5. The 
osazone of oxypyruvic acid forms lustrous, blue-black needles 
from benzene; m.p. 215 to 216°. 



1 C. Hauessermann, Z. angew. Chem., 23 (1910), 1761-1763. 

2 C. Hauesserann, Z. angew. Chem., 26 (1913), 456. 

3 C. Haeussermann, Z. Schiess-u. Sprengstoff., 2 (1907), 426. 

4 C. Haeussermann, Chem. Ztg., 29 (1905), 667. 

5 B. Rassow and W. von Bonge, Z. angew. Chem., 21 (1908), 732. 

6 C. Piest, Z. angew. Chem., 23 (1910), 1009; E. Knecht and A. Lipschitz, 
/. Soc. Chem. Ind., 33 (1914), 122. 

7 W. Will, Ber., 24 (1891), 400-407. 

8 E. Berl and W. Smith, J. Soc. Chem. Ind., 27 (1908), 537. 



326 



CHEMISTRY OF CELLULOSE AND WOOD 



Nitrates from Wood Pulp. — A nitrate of wood was prepared by 
Schultze 1 in 1865. Wood in the form of small, thin squares was 
boiled in soda solution, washed, steamed, treated with chlorine 
or bleaching powder, and nitrated with a mixture of 100 parts 
sulphuric acid and 40 parts nitric acid. This compound was 
long used as a sporting powder. Muschamp 2 cooked hardwoods 
with sodium hydroxide, obtaining what was virtually a lignin- 
free pulp, which was then nitrated. 

Guttmann, 3 in 1894, stated that wood cellulose was not used 
for explosives, since it did not seem to give as tough a powder as 
guncotton. Years later Tedesco, 4 as a result of comparative 
tests on the stability of cotton and wood pulp which had been 
nitrated with mixed acid of varying composition, came to the 
conclusion that wood pulp was too unstable for explosives. The 
cotton nitrates ignited at 181 to 190° and the wood pulp nitrates 
at 138 to 153°. On the other hand, Nitznadel 5 concluded that 
the nitrates from sulphite pulp were well suited for explosives, the 
stability tests running above the minimum limit of 180°. 
The chief disadvantages of wood-pulp nitrates in comparison 
with cotton nitrates were the higher solubility in ether-alcohol, 
lower yields, and somewhat lower ignition temperature. 

De Mosenthal 6 found the following properties for nitrated 
wood cellulose: 

Wood Cellulose Nitrates 



Number 


N, per cent 


Solubility 

alcohol-ether * 

per cent 


Density 


15 
16 
17 


13.30 
12.80 
12.50 


10 
98 
95 


1.715 
1.659 
1.673 



*Two volumes ether (sp. gr. 0.72) and one volume alcohol (sp. gr. 0.76). 

1 E. Schultze: cited by R. Escales, "Die Schiessbaumwolle," Leipzig 
(1905), p. 264. 

2 J. B. Muschamp, Ber., 5 (1872), 122. 

3 O. Guttmann, J. Soc. Chem. Ind., 13 (1894), 576. 

4 H. Tedesco, Z. Schiess-u. Sprengstoff., 7 (1912), 474-477. 

5 K. A. Nitznadel, Z. Schiess-u. Sprengstoff., 7 (1912), 257-260, 301-305 
339-343, 384-387; Wochbl. Papierfabr., 43 (1912), 3488-3489. 

6 H. de Mosenthal, J. Soc. Chem. Ind., 26 (1907), 443-450. 



THE ACTION OF ACIDS ON CELLULOSE 



327 







O B . as 




CO CO 


IO CO 






•ft 2 ft S 


O 00 GO O 


CO CO 






r£ bC 3 «* 


ICI H H CO 


TH O 






13 © ft m 


rH rH rH f-l 


i-H l-( 




a) 


PS 
















1 


B B «i 3 








"8 


•43 s g a 


>o lO rH CO 


CO CO 




00 00 00 GO 


00 GO 




a 

3 
a 


Igni 
empe 

deg 
Centi 








T3 


•+j 








o 










o 










t* 


si g 1 


h n uj ^ 


CO CO 




««H 


t> o <m o 


t- rH 




o 


£1^2 a 


o o i-i d 


r-J O 




1 


a 
















"3 


uble 
her- 
ohol, 
cent 


CD 00 CO OS 


8S 

CO lO 




B 


N "3 CO tO 
I> rH i-l CM 






£ "3 a 


i-H CO r-t 


CM 




Nitro- 
gen, 
per 
cent 


CO 00 00 id 


rH ^-1 






OS O 00 i-l 
<N CO i-l CO 


00 CM 
CM CO 






rH rH rH ^H 


i-H l-H 






Meth- 
oxyl, 
per 
cent 


IC rH O CO 


CM rH 






(O » O N 


OS rH 


CO 




oddo 


d d 


H 










« 




ento- 

san, 

per 

cent 


O CO CD IN 


rH rH 




CO rH CO CO 
i-l CO rH CO 


OS OS 
iO CO 






Ph 






jS 




















03 ♦_- 






rH 

h5 


3 


Residu 
from 

Na 2 S, 
per cen 


rH OS UO "0 


CO io 


p 


3 


io a w oo 


t^ OS 


Cm 

l 


T3 


Ol N t» N 


OS OS 


P 
O 

1 


O 

o 

* 

3 








Hydrol- 
ysis 
number 


rH CM <N i-l 


CM W5 


O) CO >o o 

OS rH (M rH 


CD CO 

d t>^ 




03 

'55 


rH rH rH 


1-1 




£> 


















3 


^ i« > 








CI 


1) 0)+^ 

ill- - !'' 


W N K3 00 


• OS 




<; 


IO i-H OS- CO 


• CM 






<Sgg 


O CO i-l tN 


: d 




•f S a < 


io m oo >o 


IO rH 






co ■* co co 


CM l-H 






3 a ® 


dodo 


rH O 




.2 as" (h +5 
o h ® S 


l> rH CO 00 


b- rH 






oo i-h os co 


OS CM 






^ -2 ft 8 


K5 N 00 N 


d t^ 












• T3 


















. CD 


















• A 


















• o 


















■ 03 


















13 £ 


















03 -° 












u 


- 0) 




T3 








ft 




Cfl 






ft 


03 


-b a 




"C 






3 


a 


o -3 




B 






P4 


a b 


§3 M 




a 








ilphite pul 

1. Nitrati 

2. Aspen. 

3. Mitsch 

4. Ritter- 


ft a 
3 .£ 


b" 
o 

1 








03 . . 
■O ICJ CD 








co 






CO 







328 CHEMISTRY OF CELLULOSE AND WOOD 

The optical rotations of the nitrates in ethylacetate and acetone 
were [a] D = +23.3 to +23.6°. Number 16 was without action 
on polarized light. 

Comparative nitrations of cotton and various wood pulps 
were carried out by Schwalbe and Schrimpff. 1 The highest 
nitrate obtained from wood pulp contained 13.34 per cent of 
nitrogen, and that from cotton, 13.46 per cent of nitrogen. A 
nitrated wood pulp could be prepared having a solubility of 5 to 
6 per cent in alcohol-ether. No difficulty was met with in 
removing the unstable impurities, so that the wood pulp nitrates 
would correspond in stability to cotton nitrates. Sulphuric 
esters were best eliminated from the nitrated product by boiling 
with 1 per cent hydrochloric acid. 

Coniferous pulps are superior to hardwood pulps owing to their 
longer fibers and lower percentage of hemicelluloses. 2 Soda 
pulps both from hardwoods and softwoods should be used in 
the unbleached condition, since bleaching does not permit them 
to fall within the government specification, that the solubility 
in boiling 10 per cent potassium hydroxide shall not exceed 
10 per cent. The presence of dark-colored products, which are 
not removable from the soda pulp by digestion with sodium 
hydroxide, results in a dark cellulose nitrate, bat this feature is 
not considered objectionable. The solubility of sulphite pulp 
in potassium hydroxide, while high originally, is not increased 
by bleaching. Previous to nitration, the sulphite pulp is purified 
by heating with a 2 per cent solution of sodium hydroxide at 
115° until the alkalinity disappears. 3 The loss in weight, which 
is about 25 per cent, is due to the removal of hemicelluloses, 
modified celluloses, lignin residues, and resin. It is then treated 
with about 4 per cent of bleach. In the form of cr6pe sheets 
both soda and sulphite pulps nitrated evenly, giving products 
with a high nitrogen content. The yield was 150 per cent, com- 
pared with 160 to 165 per cent with cotton. The wood-pulp 
nitrates were more difficult to dehydrate than the cotton 
nitrates. 

1 C. G. Schwalbe and A. Schrimpff, Z. angew. Chem., 27 (1914), 662- 
664. 

2 S. D. Wells and V. P. Edwards, Paper 25, 23 (1919), 180-185. 
3 V. P. Edwards, U. S. P. 1310694 (1919). 



THE ACTION OF ACIDS ON CELLULOSE 329 

According to Schwarz, 1 the most suitable wood pulp for nitra- 
tion is made by the Ritter-Kellner process. Wood cellulose is 
more voluminous than cotton cellulose, so that the nitrating 
charge is reduced 10 per cent. It is less adsorbent to acid, and 
the acid is not washed out as readily as from cotton. In all 
other respects, however, wood pulp is superior to cotton. 

Purification of Wood Pulps. — Sulphite pulp can be rendered 
inactive towards Fehling's solution by boiling with milk of 
lime. 2 During the treatment, half of the pentosans are destroyed. 
Lime has little action on soda pulp. This would be expected 
from the drastic alkaline treatment by which it is produced. 
Ten grams of unbleached sulphite pulp boiled with 3 grams of 
barium oxide for 3 hours showed a utilization of 13.7 per cent 
of barium oxide; and after 8 hours' boiling, a utilization of 13.4 
per cent of barium oxide based on the pulp. The loss in weight 
of the pulp was 16.2 and 17.0 per cent. Pure cotton showed no 
loss on boiling with lime water. 3 

Harding 4 heated wrapping paper — probably kraft pulp — 
with a 1 per cent solution of sodium hydroxide at 102° for 30 
hours, whereby the pulp lost 40 per cent in weight, but still 
contained lignin. As would be expected, sodium sulphide (50 
grams in 200 cubic centimeters of water) attacks sulphite pulps 
more strongly than soda pulps. Aspen, though cooked by the 
soda process, is an exception, being strongly attacked by this 
reagent. 5 

Opfermann 6 purifies wood pulp by boiling it for 3 to 6 hours 
at 2 to 3 atmospheres' pressure with a 0.5 to 2.0 per cent solution 
of sodium carbonate. The addition of a small amount of alkali 
hydroxide or sulphide increases the softness and absorbency. 
The pulp may also be purified with a solution containing 10 
grams of lime per liter. The efficiency of alkaline earth oxides 

1 R. Schwartz, Oesterr. Chem. Ztg., 22 (1919), 50-52, 57-60. 

2 C. G. Schwalbe and E. Becker, J. prakt. Chem., 100 (1919), 19-47. 
3 M. Robinoff, Diss. Darmstadt (1912). 

4 W. G. Harding, J. Phys. chem., 26 (1921), 201-203. 

5 C. G. Schwalbe, Z. angew. Chem., 31 (1918), 53. 

6 E. Opfermann, G. P. 219085 (1909); F. P. 402462 (1909); cf. C. Piest, 
Papier-Fabr., 12 (1914), 860-865. 



330 CHEMISTRY OF CELLULOSE AND WOOD 

is increased in the presence of salts of oxyacids, in which, these 
oxides are more soluble than in water. 1 

Pulps produced by alkaline processes are purified by boiling 
with a sulphite solution. 2 Pentosans and lignin can be almost 
completely removed by boiling the bleached pulp with a sulphite 
solution and again bleaching. 3 

Another method for rendering wood pulp similar to cotton, 
and suitable for cellulose nitrate and artificial silk, consists in 
treating it with oxidizing agents, such as gaseous oxides of nitro- 
gen, nitrous acid, nitric acid, perchloric acid, etc., followed by 
boiling with an alkaline solution. 4 

Drewsen 5 purifies wood pulp for nitration purposes by bleach- 
ing with 2 to 8 per cent of chlorine, then boiling with dilute 
sodium carbonate or sodium hydroxide until the pulp shows a 
solubility in potassium hydroxide of less than 7 per cent. To 
remove resins and incrusting substances, Schonlau 6 treats bleached 
wood pulp in a beating engine with a hot emulsion of water and 
turpentine. Resinous impurities on nitration, form yellow com- 
pounds which reduce the stability of the cellulose nitrate. 7 
They also prevent uniform penetration of the nitrating mixture. 
The ether-alcohol extract should not exceed 0.4 per cent. 
Sulphite pulp, freed from resinous impurities, oxycellulose, and 
hydrocellulose, by boiling with dilute alkali, gives a nitrate very 
similar to cotton nitrate. 

Straw and wood pulps high in pentosans contain, after nitration, 
correspondingly high amounts of unnitrated material. 8 The 
pentosans appear to form a coating which resists penetrance 
by the mixed acid. 

It is important that the wood pulp contain a high percentage 
of a-cellulose. That this result is attained by suitable purifica- 
tion, is shown by the following: 9 

1 C. G. Schwalbe and E. Becker, G. P. 36906 and 371507 (1919); 
E. L. Rinman, G. P. 301587 (1917). 
2 W. Schacht, G. P. 306366 (1918). 
3 C. G. Schwalbe, G. P. 378260 (1921). 
4 C. G. Schwalbe, F. P. 410460 (1909). 
5 V. Drewsen, U. S. P. 1283113 and 1283114 (1918). 
6 K. Schonlau, F. P. 469484 (1914). 

7 W. E. B. Baker, Tech. Assoc. Pulp and Paper Ind., N. Y., Feb. 6-7 
(1918), 45-47. 

8 B. Rassow and E. Dorr, J. prakt. Chem., 108 (1924), 113-186. 

9 R. G. Woodbridge, J. Ind. Eng. Chem., 12 (1920). 380-384. 



THE ACTION OF ACIDS ON CELLULOSE 



331 



Fiber 


a-Cellulose, 
per cent 


p-Cellulose, 
per cent 


7-Cellulose, 
per cent 


Purified cotton 


94.5 

95.0 
94.5 


3.0 

2.0 
2.0 


2.5 


Mildly bleached spruce sulphite 
digested 5 hours at 80 pounds 
pressure with 2 per cent NaOH 
solution 

Unbleached spruce sulphite purified 
as above 


3.0 
3.5 







Mechanical difficulties in preparing the cellulose nitrate are 
overcome by mixing the purified pulp with 50 per cent of cotton. 
This procedure has been protected by Claessen. 1 

Hydroxymethylfurfural. — When hexose carbohydrates are 
treated with the hydrogen halides in an indifferent solvent such 
as ether or chloroform, co-brommethylfurfural is obtained; it 
crystallizes in orange-yellow prisms of m.p. 59.5 to 60.5°. The 
reaction is preferably carried out by adding to the carbohydrate, 
chloroform saturated with hydrogen bromide and heating in a 
water bath for 2 hours. The corresponding chlorine derivative 
is obtained with hydrogen chloride. 2 

In the formation of the aldehyde from cellulose, the latter is 
first hydrolyzed to glucose. Hess 3 assumes that the hydrogen 

(C 6 H 10 O 5 ) n + nH 2 = nC 6 H 12 6 
CHOH - CHOH 

| | + HBr > 

CH 2 OH.CHOH CHOH .CHO 



HC- 

II 
CH 2 Br.C 



CH 

II +4H 2 0. 

C.CHO 



■O' 



bromide combines first with the glucose. This compound then 
decomposes with the elimination of three molecules of water to 
give hydroxymethylfurfural, with which the hydrogen bromide 
reacts to give co-brommethylfurfural, as shown above. 

1 C. Claessen, G. P. 300844 (1915). 

2 H. J. Fenton and M. Gostling, J. Chem. Soc, 73 (1898), 554; 75 
(1899), 423; 79 (1901), 361; 95 (1909), 1334. 
3K. Hess, Z. Elektrochem., 26 (1920), 234. 



332 



CHEMISTRY OF CELLULOSE AND WOOD 



Since fructose gave high yields of Br-methylfurfural in com- 
parison with glucose, it was assumed that the reaction was a 
criterion for carbohydrates with a ketonic structure. 1 Hibbert 2 
has recently shown that the reaction is of no value for this 
purpose, the high yields from cellulose (56 per cent) in comparison 
with glucose (12 per cent) being due largely to the absence of 
free aldehyde groups that may interfere with the reaction. 
The yields of Br-methylfurfural from 10-gram samples of various 
carbohydrates are given below: 



Material 



Yield of co-brom- 
methylfurfural 



Grams 1 Grams 2 




Swedish filter paper . 
Ordinary cotton 
Mercerized cotton . . 

Viscose 

Straw cellulose 

Lsevulose 

Dextrose 

Galactose 

Cane sugar 

a-Methylglucoside. . 
Cellobiose 



1 H. J. Fenton and M. Gostling, J. Chem. Soc, 79 (1901), 361. 

2 H. Hibbert, I.e. 

In similar manner, Erdmann 3 obtained a yield of 3.7 per cent from 
filter paper. 

Kiermayer 4 prepared a hydroxymethylfurfural, by heating 
sucrose with a solution of 0.3 per cent oxalic acid at 3 atmospheres' 
pressure for 3 hours, to which he gave the structure of /3-hydroxy- 
5-methylf urf ural ; oxime, m.p. 77 to 78°; phenylhydrazone, m.p. 
138°. Van Ekenstein 5 accepted this constitution, but later 6 

1 H. J. Fenton and M. Gostling, I.e. 

2 H. Hibbert and H. S. Hill, J. Am. Chem. Soc., 45 (1923), 176-182. 

3 E. Erdmann, Ber., 43 (1910), 2391-2398. 

4 J. Kiermayer, Chem. Ztg., 19 (1895), 1003; cf. G. Dull, Ibid., 166, 216. 

5 W. A. van Ekenstein and J. J. Blanksma, Chem. Weekblad, 6 (1909), 
217. 

6 Chem. Weekblad, 6 (1909), 1047; Ber., 43 (1910), 2355. 



THE ACTION OF ACIDS ON CELLULOSE 333 

stated that the aldehyde prepared by Kiermayer's method was 
w-hydroxymethylfurfural. Fenton had previously shown that 
his bromine compound was co-brommethylfurfural. 

HC C.OH HC CH 

II II II II 

CH 3 .C C.CHO CH 2 (OH).C C.CHO 

/3-Hydroxy-5-methylfurfural co-Hydroxymethylfurfural 

Erdmann 1 then prepared w-hydroxymethylfurfura] by treating 
co-brommethylfurfural with silver acetate. The aldehyde gave 
a semioxamazone, C 6 H 6 2 .N.NH.CO.CO.NH 2 , m.p. 216°, while 
Kiermayer's aldehyde gave a semioxamazone having no definite 
melting point, but decomposing at about 260°. It was found 
that co-hydroxymethylfurfural, owing to the labile hydroxyl 
group, could be transformed into the Kiermayer isomer by heat- 
ing with aqueous oxalic acid under pressure, or by overheating 
during its preparation. The Kiermayer aldehyde is also obtained 
in the dry distillation of cellulose. 2 

Heuser 3 obtained the following yields of hydroxymethylfur- 
fural, using a 5 per cent solution of oxalic acid at 180° for half 
an hour. In this incident, the presence of free carbonyl groups 

Hydroxy- 
methylfurfural 
Per Cent 

Cotton . 82 

Viscose 5.16 

Hydrocellulose (Knoevenagel) 10 . 84 

(hydrocellulose) improved the yield of the aldehyde. According 
to Heuser, the semioxamazone of Kiermayer's aldehyde melts at 
235 to 236°, and the phenylhydrazone at 140°, not 137° as given 4 
by Kiermayer and Erdmann. 

Aldohexoses, heated with aqueous oxalic acid, gave a yield of 
only 1.0 per cent of hydroxymethylfurfural in comparison with 
20 to 25 per cent from the ketohexoses. 6 As little as 0.001 

1 E. Erdmann, I.e. 

2 E. Erdmann and C. Schaefer, Ber., 43 (1900), 2404. 

3 E. Heuser and W. Schott, Cellulosechemie, 4 (1923), 85; cf. Ibid., 25. 

4 According to B. Tollens (Ber., 37 (1904), 303), the m.p. is 140 to 141°. 

5 W. A. van Ekenstein and J. J. Blanksma, Ber., 43 (1910), 2355. 



334 CHEMISTRY OF CELLULOSE AND WOOD 

milligram of the aldehyde can be detected by means of the red 
precipitate formed with resorcinol and concentrated hydro- 
chloric acid. 

Hydroxymethylfurfural is apparently an intermediate product 
in the formation of lsevulinic acid from carbohydrates. 

HC C.OH 

II II 2H 2 
CH 3 .C C.CHO > CH 3 .CO.CH 2 .CH 2 .COOH + H.COOH. 

V 

It reduces Fehling's solution readily, and to twice the extent 
of glucose. 



CHAPTER X 
SACCHARIFICATION OF CELLULOSE AND WOOD 

Saccharification is of more narrow scope than hydrolysis and 
refers to the conversion of cellulose to the simple sugars. In the 
case of "sulpholysis", hydrolysis and formation of dextrin 
esters of sulphuric acid take place simultaneously. Saccharifica- 
tion always appears to take place through the intermediate 
formation of dextrins. 

Saccharification of Cellulose with Concentrated Acids. — In 
1819, Braconnot 1 dissolved linen in sulphuric acid of sp. gr. 1.827. 
After standing some time, the solution was diluted, boiled, the 
acid removed with lime, and the sugar solution carried to dryness. 
In this way 23.3 grams of sugar were obtained from 20.4 grams 
of linen, so that the yield presumably was quantitative. A 
crystalline, fermentable sugar identical with glucose was isolated. 

Numerous subsequent attempts have been made to obtain 
a quantitative conversion of cellulose to glucose. Flechsig 2 
dissolved cotton in 69 per cent sulphuric acid, let the solution 
stand 45 minutes, diluted to an acid concentration of 2.5 per 
cent, and boiled for 5 to 6 hours. Reduction with Fehling's 
solution showed 96.9 per cent of glucose, based on the cellulose. 
Crystalline glucose was obtained, though the yield was not 
quantitatively determined. Ost and Wilkening 3 confirmed 
Flechsig' s results. They obtained from 90 to 95 per cent of the 
theoretical yield of sugar as determined by the cupric reducing 
value. Fermentation showed 80 to 83 per cent of the apparent 
dextrose values. 

In repeating Flechsig' s work, Schwalbe and Schulz 4 threw 
doubt on the high yields of sugar reported, it having been found 

1 H. Braconnot, Ann. chim. phys., [2] 12 (1819), 172; 25 (1827), 81. 

2 E. Flechsig, Z. physiol. Chem., 7 (1883), 528. 

3 H. Ost and L. Wilkening, Chem. Ztg., 34 (1910), 461. 

4 C. G. Schwalbe and W. Schulz, Ber., 43 (1910), 916. 

335 



336 CHEMISTRY OF CELLULOSE AND WOOD 

that the reducing value was not a reliable indication of the 
amount of glucose . present. An appreciably higher reduction 
was obtained when the acid was neutralized with barium carbon- 
ate than with sodium hydroxide. In some cases a high reducing 
value was apparently due in considerable part to dextrins; in 
others, a high yield of sugar could be obtained, though the reducing 
value was low. Artificial mixtures of Guignet's soluble cellulose 
and glucose showed abnormally low reducing values. Storer 1 
had 10 years previously called attention to the error involved in 
assuming that in hydrolyzing cellulose the reducing value could 
be quantitatively referred to glucose. 

Muhlmeister 2 used Flechsig's method on various celluloses. 
Using 70 per cent sulphuric acid, cotton gave 104.8 per cent of 
glucose by reduction, and 79 per cent of the theoretical yield of 
alcohol; Ritter-Kellner sulphite pulp gave 90.6 per cent of glucose 
by reduction, and 83.1 per cent of the theoretical yield of alcohol. 
By a similar procedure, Neuman 3 obtained 87 to 91 per cent of 
glucose as determined by fermentation and the polariscope. 

The results of Monier- Williams 4 leave no doubt of a possible 
quantitative saccharification of cellulose, since he obtained 90.67 
per cent of the theoretical amount of crystalline glucose. Ten 
grams of cotton, having 6.93 per cent moisture and 0.13 per cent 
ash, were dissolved in 50 cubic centimeters of 72 per cent sulphuric 
acid and allowed to remain at room temperature for 1 week. 
The acid was diluted to 5 liters and boiled under a reflux con- 
denser for 15 hours. The liquid was filtered, neutralized with 
barium carbonate, filtered, and the filtrate evaporated to dryness 
at reduced pressure (40 millimeters). To take care of the 
alkalinity that developed during evaporation, a few drops of 
methyl red were added and the solution kept neutral by the 
repeated addition of 0.1N sulphuric acid. The residue was 
extracted with methyl alcohol free from acetone. After filtra- 
tion and decolorization with a little animal charcoal, the solvent 

1 F. H. Storer, Bull Bussey Inst., 2 (1900), 463. 

2 H. Muhlmeister, "Beitrage zur Hydrolyse und Sulfolyse der Zellulose," 
Diss. Hannover (1913), p. 11. 

3 J. Neuman, "Kritische Studien iiber Hydrolyse der Cellulose und des 
Holzes," Dresden (1910), p. 34. 

4 G. W. Monier- Williams, J. Chem. Soc, 119 (1921), 803-805. 



SACCHARIFICATION OF CELLULOSE AND WOOD 337 

was evaporated in a current of dry air at a low temperature. 
There were obtained 9.718 grams of nearly white crystals. 

Sulphuric acid of 70 per cent concentration is, according to 
Ost, 1 more suitable for the sulpholysis of cellulose than a stronger 
acid. It forms dextrin esters of sulphuric acid which on dilution 
and subsequent heating at 120° may be almost quantitatively 
hydrolyzed to glucose. Any "amyloid" formed on dilution 
must be redissolved in strong acid before it can be hydrolyzed. 

Strong sulphuric acid in time causes blackening of the carbo- 
hydrates due to dehydration. Tschumanow 2 in this way pre- 
pared colloidal "carbon." Strips of filter paper were left in 
contact with concentrated sulphuric acid for 3 days and the 
dark-colored liquid obtained was diluted with a large volume of 
water. The precipitate, dried at 100 to 112°, contained 62.1 per 
cent C, 5.0 per cent H, and 32.9 per cent O. This corresponds 
with the formula (C 2 H 2 0) n and approaches ulmic acid in composi- 
tion. When 2.59 grams of cellulose were dissolved in 100 cubic 
centimeters of concentrated acid and let stand 2^ years, a black, 
quivering, gelatinous mass was obtained. In a similar manner, 
gelatinous carbon and a stable colloidal carbon have been 
prepared from sugar. 3 

In 1880, Dauziville 4 described a method of hydrolysis in which 
moist cellulose was treated with hydrogen chloride gas until 
swelling and solution took place. After standing to further the 
hydrolysis, the hydrochloric acid was removed for the most part 
by suction, the residue strongly diluted, and boiled to produce 
sugar. 

It was observed by Willstatter and Zechmeister 5 that hydro- 
chloric acid of a specific gravity greater than 1.20 and containing 
40 to 42 per cent of hydrogen chloride rapidly dissolved cellulose. 
When 1 gram of cotton was dissolved in 100 cubic centimeters 
of acid of sp. gr. 1.204 and let stand at 15 to 16° for 22 hours, 
the reducing value indicated 95.4 per cent, and the optical rotation 
96.3 per cent, of the theoretical yield of glucose. 

1 H. Ost, Ann., 398 (1913), 320. 

2 S. Tschumanow, Kolloid-Z., 14 (1914), 321-322. 

3 L. Sabbatini, Kolloid-Z., 14 (1914), 29-31. 

4 E. S. Dauziville, G. P. 11836 (1880). 

5 R. Willstatter and L. Zechmeister, Ber 46 (1913), 2995-2998; 
L. Zechmeister, Diss. Zurich (1913), p. 37. 



338 CHEMISTRY OF CELLULOSE AND WOOD 

The rotatory power of glucose increases greatly with the 
strength of the acid. The following values were observed for 1 
gram of glucose in 100 cubic centimeters of acid at 16 to 17°: 



d 15°/4° 


Per Cent 
HC1 


w* 


1.018 


3.68 


+54.5' 


1.154 


30.4 


61.0 


1.204 


39.9 


97.5 




44.5 


164.6 



The increased rotation was attributed to the formation of iso- 
maltose. According to Daish, 1 the high rotation of dextrose in 
fuming hydrochloric acid is due to almost complete transforma- 
tion of the sugar into the a-form, whose rotation is of the order 
[«]z> = +106°. The course of the hydrolysis of cotton cellulose 
was followed by observing the rotation at intervals. 2 After 
40 minutes, [a] D was +1-0°, this value increasing to and becoming 
constant at 98.5° after 3100 minutes. When the rotation was 
plotted against time, a sharp break in the curve was noted between 
the second and the third hour. 

The work of Willstatter and Zechmeister has been criticized 
by various investigators. Ost 3 objected to the claim of novelty 
in the use of hydrochloric acid and asserted that satisfactory 
evidence for quantitative saccharification cannot be obtained by 
the use of either Fehling's solution or the polariscope. Cunning- 
ham 4 showed that, contrary to previous opinion, hydrochloric 
acid forms esters with the degradation products of cellulose. 
It was also observed that celluloses varying so widely in com- 
position as cotton and esparto gave identical rotations. In 40 
per cent hydrochloric acid, cotton showed [a] D = +56.8° after 
2250 minutes; esparto cellulose, [a] D = +46.3° after 2250 
minutes, increasing to +54.4° after 3000 minutes. Slight 
variations in the concentration of the hydrochloric acid cause 
marked constitutional changes in the simple hexoses, as shown 
by the marked fluctuations in optical activity. 

1 A. J. Daish, J. Chem. Soc., 106 (1914), 2053. 
2 L. Zechmeister, I.e. 

3 H. Ost, Ber., 46 (1913), 2995. 

4 M. Cunningham, J. Chem. Soc., 113 (1918), 173-181. 



SACCHARIFICATION OF CELLULOSE AND WOOD 



339 



It is not apparent how celluloses containing pentosans can 
give the same rotation as cotton. The xylan in esparto cellulose 
is changed to furfural in hydrochloric acid of high concentration. 1 

Cotton cellulose, and the celluloses of yellow birch, white 
spruce, and Douglas fir, isolated by chlorination, when dissolved 

no 



100 













































































V > 


























>: 




i 














































3 
or 


7/ 


t 






























































































// 


J Ye 

1 


How Birch: 
OOOgm.inlOOc 


c.HCl Sp.gr. HCt =1.205 










Y 


Douglas Ffr; 
I.OOOgm. inlOOc.c.HCl. 5p.gr. HCI-- 1.205 












CoHon: 
/. 97l4gm. m200c.c.HCL Sp.gr.HCl= 1.202 












White Spruce: 























































































) 


A 




( 




8 




1 


3 


i; 


5 


14 



90 



80 



70 



GO 



50 



40 



30 

Ti'me in Hours 
Fig. 7. — Change in specific rotation with time of hydrochloric acid solutions of 
various celluloses. (After Sherrard and Froehlke.) 

in 41 per cent hydrochloric acid, were found by Sherrard and 
Froehlke 2 to give similar but not identical optical rotation 
curves (Fig. 7). In addition to the break in the curves between 
the second and third hours, another break occurred at about the 
seventh hour. Assuming that glucose is the only sugar formed 
during the hydrolysis, the amounts of sugar formed are practi- 

1 E. L. Hirst and D. R. Morrison, J. Chem. Soc, 123 (1923), 3226. 

2 E. C. Sherrard and A. W. Froehlke, J. Am. Chem. Soc, 46 (1923), 
1729-1734. 



340 



CHEMISTRY OF CELLULOSE AND WOOD 



cally identical for the four celluloses, namely, 91.40 per cent for 
cotton to 90.10 per cent for birch cellulose. This is remarkable 
in view of the presence of mannan in spruce cellulose and of 
about 28 per cent of pentosan in birch cellulose. 

Cellulose, when dissolved in a solution of one part of zinc 
chloride in two parts of concentrated hydrochloric acid, reaches 
a maximum rotation and then decreases. 1 No breaks in the 
curves were obtained as with hydrochloric acid alone. The 
different celluloses showed a considerable variation in rotation. 
In the solvent employed, the specific rotation for pure glucose 
was +60.0°, and for xylose +32.0°. The higher rotation of 
cotton cellulose was attributed to the formation of dextrins. 

Optical Rotation of Cellulose in Zinc Chloride-Hydrochloric Acid 



Kind of cellulose 



Method of 
isolation 



xime, 
hours 



Maximum 
specific 
rotation 



Cotton 

Cotton 

Cotton 

Jute 

Jute 

Hydrocellulose 
Oxycellulose . . . 

Fir 

Beech 

Beech 



J. Konig 

B. Tollens 

Cross and Bevan 

J. Konig 

B. Tollens 

Girard 

Vignon 

J. Konig 

J. Konig 

H. Muller 



72.5 

78 

55 

56.5 

49 

24 

30 

71.5 

38 

24 



+82.55° 
72.02 
56.72 
71.50 
71.57 
45.02 (?) 
52.29 
60.51 
62.00 
48.84 



Krull 2 found that 97 per cent of the theoretical yield of glucose 
could be obtained from cotton by hydrolysis with strong hydro- 
chloric acid. The glucose present was determined with Fehling's 
solution. The cellulose, containing three parts of water, was 
treated with hydrogen chloride gas until saturated. Hydrolysis 
was allowed to proceed for 5 to 18 hours at 20°. By means of a 
vacuum, the hydrochloric acid content was then reduced to 2 
per cent. The solution, reduced to a cellulose content of 1 to 10 



1 J. Konig and F. Huhn, "Bestimmung der Zellulose in Holzarten" 
(1912), 55. 

2 H. Krull, "Versuche iiber Verzuckerung von Zellulose," Diss. Danzig 
(1916), 62 pp. 



SACCHARIFICATION OF CELLULOSE AND WOOD 



341 



per cent by dilution with water, was boiled for 8 to 10 hours to 
complete the hydrolysis. (Krull states that the data given in 
the table below were obtained by means of a primary hydrolysis 
lasting 20 hours, followed by diluting and boiling for 8 hours; in 
another paper 1 where the same data are given, it is stated that the 



Material 


Sugar, 
per cent 


Alcohol, 
per cent 


Residue, 
per cent 


Cotton 

Cotton 


94.39 

97.86 

13.2 

84.9 

70.43 

88.65 


46.60 

44.28 

Qual. 

29.98 

14.0 

25.34 




Hydrocellulose (Girard) 


77.28 


Sulphite pulp, unbleached 

Sulphite pulp, strongly bleached 

Soda pulp, unbleached 


12.84 



primary hydrolysis lasted 5 hours.) On extending the period of 
primary hydrolysis to 24 hours or beyond, "reversion" took 
place with the formation of relatively large amounts of non- 
fermentable products. 

The cellulose, according to Kauko, 2 should contain two parts 
of water previous to saturation with hydrogen chloride. A larger 
amount of water reduces the yield of dextrose. At 13°, an 80 
per cent yield of sugar can be obtained in 4 hours, and 90 per 
cent, in 40 hours. Vapor-tension measurements indicate that 
cellulose forms an addition product with hydrochloric acid, two 
mols of hydrogen chloride combining with each C 6 group. 

Cotton in ordinary concentrated hydrochloric acid, after 40 
hours' action, does not reduce Fehling's solution but does so 
after 87 hours. 3 

Boiling solutions of hydrofluoric acid, of concentrations of 0.5 
to 30.0 per cent, act very slightly on cellulose. At the tempera- 
ture of boiling water, acid of 40 to 50 per cent concentration gave 
in 6 hours, 41 per cent of glucose based on the weight of the 
cellulose. Further heating caused decomposition of the sugar. 4 

Saccharification of Cellulose with Dilute Acids. — The yield 
of dextrose by the hydrolysis of cellulose with dilute acids under 

1 A. Wohl and H. Krull, Cellulosechemie, 2 (1921), 7. 

2 Y. Kauko, Naturwissenschaften, 9 (1921), 237-238. 
3 W. O. de Coninck, C. A., 5 (1911), 1585. 

4 J. Ville and W. Mestrezat, Compt. rend., 150 (1910), 783-784. 



342 



CHEMISTRY OF CELLULOSE AND WOOD 



pressure, is much lower than that obtained by the use of strong 
acids in the cold. A subsequent hydrolysis of the residue yields 
more sugar, but the yield is much reduced over the first. The 
low yield of sugar is due in part to the destruction of some of 
the sugar formed. 

Decomposition of sugar by acids produces for the most part 
formic and lsevulinic acids, and humus substances. The destruc- 
tive action of dilute sulphuric acid, under pressure, on glucose, is 
shown in the following table. 1 



Dextrose used 


Heated with 2.5 per 
cent H2SO4 


Dextrose 

decomposed, 

grams 


Found, referred to 100 grams 
of decomposed glucose 


(in 5 per cent 
solution), grams 


Temperature 


Time, 
hours 


Formic 
acid, 
grams 


Laevulinic 
acid, 
grams 


Humus, 
grams 


20 
40 
20 
40 
40 
40 


145 
145 
145 
145 
120 
120 


7 

5 

3 

1.5 

8 

3 


15 
26 

7.5 
13 
11 

4 


16.2 
17.4 
13.4 
17.8 
16.9 
8.6 


32.4 
40.2 
30.3 

32.2 

44.0 

8.1 


13.3 

12.3 

8.0 

8.5 

8.2 




The following results were obtained by Neuman: 2 

Decomposition of 1 Gram of Glucose in 25 Cubic Centimeters of 

Sulphuric Acid of Various Concentrations Heated for Half 

an Hour at the Corresponding Temperature 



H2S04, 


Per cent undecomposed glucose at T = 


per cent 


150° 


160° 


175° 


185° 


0.1 


100 


94.4 


94.2 


88.8 


0.5 


96.1 


92.7 


91.6 


50.0 


1.0 


94.4 


83.3 


86.6 


33.3 


1.5 


88.8 


80.5 


55.5 


31.1 


2.0 


87.7 


75.0 


37.2 


5.5 


2.5 


86.6 


72.2 


33.3 


5.0 


3.0 


83.3 


71.0 


25.0 


2.7 


5.0 


80.5 


38.8 


5.5 


0.0 



1 H. Ost and T. Brodtkorb, Chem. Ztg. 
L. Wilkening, Ibid., 34 (1910), 461. 

2 J. Neuman, I.e., 31. 



35 (1911), 1126; cf. H. Ost and 



SACCHARIFICATION OF CELLULOSE AND WOOD 343 

In the hydrolysis of cellulose a temperature of 175° should not 
be exceeded, since this is near the " critical point" for the decom- 
position of glucose. The decomposition of mannose is similar 
to that of glucose. 1 

Cotton, hydrolyzed six times in succession with 1 per cent 
sulphuric acid at 150° for 2 hours, gave a total reducing value 
equivalent to only 44.3 per cent of glucose. 2 Muhlmeister, 3 
under the same conditions, obtained 12.3 and 8.7 per cent of 
glucose by two hydrolyses, making a total of 21.0 per cent. 
Two cooks with 2 per cent acid gave a total of 33.6 per cent of 
glucose. Neuman 4 treated cotton with 0.5 per cent sulphuric 
acid at 175° for 30 minutes, and repeated the hydrolysis on the 
undissolved portion: 

1. Cotton 26 . 6 per cent glucose (on original) 

2. Residue from (1) 15.0 per cent glucose (on residue) 

3. Residue from (2) 12.0 per cent glucose (on residue) 

4. Residue from (3) 13.3 per cent glucose (on residue) 

The total yield of glucose by the four hydrolyses was 46.9 per 
cent based on the original cellulose. 

The influence of pressure, time, temperature, and amount and 
concentration of sulphuric acid on the hydrolysis of sulphite 
pulp was studied by Simonsen. 5 Pressure, as obtained with 
chloroform, e.g., is of no purpose without a correspondingly high 
temperature. In the tables given on page 344, the temperature 
corresponds with the various steam pressures. 

Simonsen 6 concluded that from the standpoint of economy of 
operation the most suitable conditions for hydrolysis were: 
one part of cellulose and six parts of 0.5 per cent sulphuric acid 
heated for 1.5 hours at a pressure of 10 atmospheres. The yield 
of reducing sugars was 41 per cent. The yield of sugar is influ- 
enced by two principal factors, the speed of sugar formation 
and the speed of destruction of the sugar. It has been found 

l E. C. Sherrard and W. H. Gauger, /. Ind. Eng. Chem., 15 (1923), 
1164. 

2 H. Ost and L. Wilkening, Chem. Ztg., 34 (1910), 461. 

3 H. Muhlmeister, I.e., 38. 

4 J. Neuman, I.e., 40. 

5 E. Simonsen, Z. angew. Chem., 11 (1898), 219-228. 
6 E. Simonsen, I.e., 223. 



344 



CHEMISTRY OF CELLULOSE AND WOOD 



Influence of Pressure and Concentration of Acid 
(40 Grams Cellulose and 1080 Cubic Centimeters of Dilute H 2 S0 4 . Time 

4 Hours) 



Pressure, 
atmospheres 



Concentration of acid, per cent 



0.15 per cent 0.30 per cent 0.45 per cent 0.6 per cent 



Yield of sugar based on cellulose 



1.3 




2.5 


2.7 


3.1 


2.1 




6.6 


8.6 


10.6 


2.7 




9.3 


11.3 


12.6 


4.0 




16.4 




20.3 


6.0 


21.5 


28.0 


30.7 


43.9 


8.0 


30.5 


38.4 


45.0 


33.3 


9.0 




43.1 






10.0 


35.0 


36.6 


30.0 


18.0 


12.0 


38.4 








14.0 


20.0 









Influence of Amount of Acid 

(40 Grams Cellulose and 4.8 Grams H 2 S0 4 in Following Amounts of Water. 

Time, 2 Hours; Pressure, 8 Atmospheres) 



Water, grams 


Concentration of acid, 
per cent 


Yield of sugar based 
on cellulose, per cent 


250 

500 

1,080 

1.500 


1.5 
0.75 
0.44 
0.23 


18.4 
35.0 
45.4 
32.5 



Influence of Time 
(40 Grams Cellulose and 250 Cubic Centimeters 0.5 Per Cent H 2 S0 4 . Pres- 
sure, 10 Atmospheres) 



Time, minutes 


Yield of sugar based on cellulose, 
per cent 


15 


20.1 


37 


37.3 


45 


38.2 


60 


39.0 


75 


40.0 


90 


42.7 


120 


35-40.9 



SACCHARIFICATION OF CELLULOSE AND WOOD 345 

preferable, therefore, in the case of wood to bring the pressure 
to 7.5 atmospheres as rapidly as possible and then relieve it. 

Hydrolysis of cotton with sulphurous acid by the Classen 
process gave only 6.4 per cent of glucose. 1 

Cellulose, on exposure to the vapors of hydrochloric acid, is 
partially hydrolyzed to glucose. Air-dry cotton, after exposure 
for 1 week, gave 10.6 per cent of glucose, and after 2 weeks' 
exposure, 18.0 per cent. 2 The insoluble residue from the latter 
experiment, on hydrolysis with 0.5 per cent sulphuric acid at 
175° for 30 minutes, gave 40 per cent of glucose, i.e., 32 per cent 
based on the original cellulose. This shows that, contrary to 
the statements of others, Girard's hydrocellulose shows no 
peculiar resistance to hydrolysis. 

Glucose, 3 identified through the osazone, is also formed in the 
preparation of hydrocellulose by treating cotton with dilute 
sulphuric acid, drying, and heating at 70°. 

Sugars and Ethyl Alcohol from Wood. — Processes for the pro- 
duction of fermentable sugars from wood involve the use of: (1) 
dilute acids at high temperatures; (2) concentrated acids at low 
temperatures. Neither class has proved to be an unqualified 
success. The yield of fermentable sugars with dilute acids is 
insufficient to give a financially comfortable margin of safety, 
except under special circumstances. Concentrated acids give 
a high yield of sugars, but the recovery of the acids, which is 
necessary for economical operation, is beset with many engineer- 
ing difficulties and requires a high plant investment. 

Saccharification with Dilute Acids. — The hydrolysis of cellu- 
lose to glucose, though discovered in 1819, received scant atten- 
tion until the middle of the century, when numerous papers of 
an industrial and scientific nature were published. Melsens 4 
appears to have first hydrolyzed wood in an autoclave with dilute 
acid. He used concentrations of sulphuric acid of 3 to 10 per 
cent, and temperatures up to 200°. Payen 5 proposed to treat 

1 J. Neuman, I.e., 48. 

2 J. Neuman, I.e., 66. 

3 W. Netthofel, "Beitrage zur Kenntnis der Zellulose," Diss. Berlin 
(1914), p. 36. 

4 G. F. Melsens, Dinglers poly tech. J., 138 (1855), 426-429. 

5 A. Payen, Compt. rend., 64 (1867), 1167-1174. 



346 



CHEMISTRY OF CELLULOSE AND WOOD 



wood with dilute hydrochloric acid to obtain pulp, sugars being 
a by-product. Experiments by Thorn 1 gave 18 to 25 per cent 
of sugars. Zetterlund 2 heated wood with dilute hydrochloric 
acid at a pressure of 0.12 kilogram per square centimeter and 
obtained 19.7 per cent of sugars which were fermented to alcohol. 
He thought that hardwoods gave the best results. 

The first systematic investigation was made by Simonsen, 3 
who studied the effect of time, temperature, and concentration 
of acid. He heated 100 grams of sawdust with 500 cubic centi- 
meters of sulphuric acid and obtained the following yields of 
sugar: 



Effect of strength of acid 
(1 hour; 10 atmospheres) 


Effect of pressure 

(temperature) (1 hour; 

0.5 per cent acid) 


Effect of time 

(9 atmospheres; 0.5 

per cent acid) 


Strength of 
acid, per cent 


Yield of 

sugar, 

per cent 


Pressure, 
atmos- 
pheres 


Yield of 

sugar, 

per cent 


Time, 
minutes 


Yield of 

sugar, 

per cent 


0.30 
0.58 
0.75 
1.00 
1.50 


18.4 
19.6 
19.2 
19.6 
5.0 


8 

9 

10 

11 

12 


18.2 
19.6 
19.6 
19.2 
15.2 


7.5 

15 

30 

60 

90 


22.2 
22.5 

20.8 
19.6 
16.6 



He concluded that optimum conditions for saccharification were : 
0.5 per cent acid; 9 atmospheres' pressure; 15 minutes' heating; 
and a ratio of dilute acid to wood of 5:1. The ratio could be 
reduced to 2.5:1 with hydrochloric acid, without decreasing the 
yield. Cooking of the residual wood gave 13 per cent additional 
sugar, a yield which did not justify recooking. Fermentations 
made on a semicommercial scale gave 2 to 6 per cent of alcohol, 
based on the dry wood. 



1 W. Thorn, Dingier s polytech. J., 210 (1873), 37-39. 

2 C. G. Zetterlund, Wagner's Jahresber., 18 (1872), 597. 

3 E. Simonsen, Biedermann' s Zentr., 25 (1896), 47-50; Z. angew. Chem., 
9 (1898), 195-196, 219-228, 962-966, 1007-1012; U. S. P. 607091 (1898). 



SACCHARIFICATION OF CELLULOSE AND WOOD 347 

Reiferscheidt 1 and Korner 2 repeated Simonsen's work and 
likewise obtained 6 per cent of alcohol. Neuman 3 concluded 
that optimum conditions were 0.5 per cent acid, a temperature 
of 175°, and a reaction period of 30 minutes. A ratio of acid to 
wood of 3:1 was as favorable as 20:1. He obtained about 20 
per cent of sugar. It was possible to obtain 44 per cent of sugar, 
or 16.6 per cent of alcohol, by hydrolyzing the wood several 
times. 

Voerkelius 4 used sulphuric acid of a concentration of 0.2 to 
2.0 per cent, at temperatures of 120 to 180°, and obtained uniformly 
about 25 per cent of sugar, or 8.6 per cent of alcohol. 

The use of suphur dioxide was patented by Classen. 5 Among 
numerous defects that prevented the success of the process, the 
most important were the large amount of acid required and the 
long conversion time (4 to 6 hours). It was found that the more 
rapidly the contents of the digester could be brought to the 
required temperature, the shorter was the time required to obtain 
a maximum yield of sugar. 6 The conversion period was reduced 
to 45 minutes by the process of Ewen and Tomlinson, 7 in which 
live steam and sulphur dioxide were introduced directly into 
the digester. Sulphur dioxide was subsequently abandoned and 
dilute sulphuric acid substituted. 8 The operation of the process 
in a plant at Georgetown, S. C, has been described by von 
Demuth 9 and Foth. 10 

A distinct contribution to the art was made by Cohoe. 11 He 
heated the sawdust with live steam in such a manner that the 

1 Reiferscheidt, Z. angew. Chem., 18 (1905), 44-48. 

2 T. Korner, "Zur Frage der Bildung von Alkohol aus Cellulose und 
cellulosehaltigen Stoffen," Diss. Dresden (1907); Z. angew. Chem., 21 
(1908), 2353-2359. 

3 J. Neuman, "Kritische Studien uber Hydrolyse der Cellulose und des 
Holzes," Diss. Dresden (1910), 80 pp. 

4 G. A. Voerkelius, Wochbl. Papierfabr., 42 (1911), 852-855. 

5 A. Classen, G. P. 118540 (1899); 130980 (1901). 

6 R. F. Ruttan, J. Soc. Chem. Ind., 28 (1909), 1290-1294. 

7 M. F. Ewen and G. H. Tomlinson, U. S. P. 763472 (1904). 

8 M. F. Ewen and G. H. Tomlinson, U. S. P. 938308 (1909). 

9 R. von Demuth, Z. angew. Chem., 26 (1913), 786-792. 

10 G. Foth, Chem. Ztg., 37 (1913), 1221-1222, 1297-1298. 

11 W. P. Cohoe, J. Soc. Chem. Ind., 31 (1912), 513-515; U. S. P. 985725 
and 985726 (1911). 



348 CHEMISTRY OF CELLULOSE AND WOOD 

amount of water present did not exceed 50 per cent of the weight 
of the sawdust, and when the gage pressure reached 50 pounds, 
hydrochloric acid was introduced. In 3 to 4 minutes, a maximum 
yield of 20 per cent of sugars was obtained. This "dry cook" 
has been generally adopted. The preliminary steaming of hard- 
woods gave 1 to 2 per cent of acetic acid. 

Ungar 1 studied the saccharification of wood with both con- 
centrated and dilute acids, but obtained yields no higher than 
his predecessors. Hagglund 2 used concentrations of sulphuric 
acid of 0.25 to 1.5 per cent. He found that the best conditions 
were a temperature of 175°, and a ratio of wood to acid of 1:3. 
The highest yield, 20.8 per cent of sugar, or 87 liters of pure alco- 
hol per long ton, was obtained with a 0.7 per cent acid and a 
conversion period of 30 minutes. 

The most extensive investigation involving the use of dilute 
acids has been made by Kressmann. 3 The optimum conditions 
for coniferous woods were: 

1. The temperature and pressure should not exceed 7.5 
atmospheres. 

2. This temperature and pressure should be reached as soon 
as technically possible, that is, in 15 to 20 minutes. 

3. The digester should be held at the above temperature and 
pressure for 15 to 20 minutes. 

4. The ratio of water to dry wood should be about 1.25: 1.0. 

5. The ratio of sulphuric acid (100 per cent) to dry wood should 
be from 1.8 to 2.5 to 100. 

6. At the end of the cooking period, the digester should be 
blown as quickly as possible. 

Under the above conditions, a yield of 25 gallons of 190-proof 
alcohol per dry ton of wood can be obtained, allowance being 
made for 2 to 5 per cent loss in distillation. Broad-leaved trees 
give only one-half as much fermentable sugars as do conifers. 

Kressmann used concentrations of sulphuric acid up to 5 
per cent of the weight of the wood. More concentrated acids 

1 E. Ungar, Diss. Budapest (1916). 

2 E. Hagglund, J. prakt. Chem., 91 (1915), 358-364. 

3 F. W. Kressmann, "The Manufacture of Ethyl Alcohol from Wood 
Waste," 11. S. Dept. Agr. Bull, 983 (1922), 100 pp.; J. Ind. Eng. Chem., 6 
(1914), 625-630; 7 (1915), 920-923; Met. Chem. Eng., 15 (1916), 78-82. 



SACCHARIFICATION OF CELLULOSE AND WOOD 349 

will give more sugar, but not enough to justify the increased cost 
of the acid. Purvis 1 heated 25 grams of sawdust, 300 cubic 
centimeters of water, and 50 cubic centimeters of 85 per cent 
sulphuric acid, just below the boiling point, with constant stirring, 
for 5.25 hours. He obtained 39 per cent of sugars. 

A high yield of reducing sugars does not necessarily mean a 
high yield of alcohol, so that the sugars should always be fer- 
mented. Sherrard and Gauger 2 have used sulphuric acid up to 
10 per cent of the weight of the wood. With 7.5 per cent of 
acid, at a pressure of 120 pounds for 15 minutes, they obtained 
only 23.6 per cent of reducing sugars, but the yield of alcohol 
was very high, namely, 29 gallons of 190-proof alcohol per ton 
of wood. 

The actual yields of alcohol in plant operations have always 
fallen far below those obtained on a semicommercial scale. 
Tomlinson 3 states that the goal to be aimed at is 20 to 22 per 
cent of fermentable sugars or 35 gallons of alcohol per ton. 
The Georgetown plant is stated to have attained a yield of 19 
gallons. Trial operations at the plant at Fullerton, La., in 
1913, were exceedingly disappointing, 4 but during the war the 
yield was brought up to 15 gallons. There is every reason to 
believe that a properly constructed plant would produce 20 
gallons of 95 per cent alcohol per ton of dry wood. 5 

Attempts have been made to increase the yield of sugars by 
the use of modified celluloses. Ekstrom 6 dissolved cellulose in 
70 per cent sulphuric acid and after a few minutes precipitated 
it as an acid cellulose by the addition of water. The modified 
cellulose was not affected by dilute acids at atmospheric pressure, 
but was converted entirely into glucose at higher pressures. 
The acid cellulose was converted to soluble cellulose by triturat- 
ing and heating it with one-fourth to one-half of its weight of 
70 to 77 per cent sulphuric acid, at a temperature of about 80°; 

1 J. E. Purvis, Proc. Cambridge Phil Soc, 19 (1919), 259-260. 

2 E. C. Sherrard and W. H. Gauger, J. Ind. Eng. Chem., 15 (1923), 1165. 

3 G. H. Tomlinson, "Manufacture of Ethyl Alcohol from Wood Waste," 
Hon. Adv. Council for Sci. Ind. Res., Canada, Bull. 7 (1919), 9 pp. 

4 G. H. Tomlinson, J. Ind. Eng. Chem., 10 (1918), 859-861. 

5 E. C. Sherrard, Chem. Age, 29 (1921), 76-79. 
6 G. Ekstrom, G. P. 193112 (1906). 



350 CHEMISTRY OF CELLULOSE AND WOOD 

it was then saccharified by diluting with water and boiling at 
atmospheric pressure for 0.5 to 1.0 hour. 1 Another method 
involved treating wood with sulphuric acid and sodium sulphate 
to produce amyloid, which was subsequently hydrolyzed with 
dilute acids. 2 

Schwalbe 3 proposes to treat wood containing less than 50 
per cent of water with gaseous acids, such as sulphur dioxide 
and hydrogen chloride, until it swells but does not dissolve. 
The wood is then saccharified by heating with dilute acids. A 
nearly theoretical yield of sugars is claimed. 

Gelatinization or swelling by means of acids, alkalis, and salts 
facilitates hydrolysis due to greater surface exposure ; mechanical 
gelatinization is without effect. The formation of hydrocellu- 
lose should be avoided. The fact that cellulose can be only 
partially converted into glucose by means of dilute acids at 
atmospheric or higher pressures, and that the residue is more 
resistant to hydrolysis than the original, has never been satis- 
factorily explained; this phase of the subject has been discussed 
in the section on hydrocellulose. The data are conflicting and 
more work is needed. Neither the "reversion" hypothesis or 
that of a "skin effect," wherein the layer of hydrocellulose resists 
penetration by the acid, is fully acceptable. The saccharification 
of cotton is influenced but little by the presence of added glucose. 

Much work has been done to increase the amount of sugar by 
the addition of catalysts other than the main acid. Gentzen 
and Roth 4 used ozone and claimed a yield of 34 per cent of glu- 
cose. Korner 5 found that potassium dichromate and potassium 
persulphate decreased the yield of alcohol; however, when a 2 
per cent solution of hydrogen peroxide was used in place of water 
for dissolving the sulphuric acid, the yield of alcohol was nearly 
doubled, wood giving 9.0 to 10.7 per cent. Treatment of wood 
with ozone before hydrolysis decreased the yield. 

1 G. Ekstrom, G. P. 207354 (1907). 

2 M. Mendelsohn and E. Frankl, G. P. 220634 (1908). 

3 C. G. Schwalbe, G. P. 305690 (1916); Zellstoff u. Papier, 4 (1924), 
115-117. 
4 R. Gentzen and L. Roth, G. P. 745675 (1903). 
6 T. Korner, I.e. 



SACCHARIFICATION OF CELLULOSE AND WOOD 351 

Krull 1 treated wood with ozone, atmospheric oxygen at 150 
to 170° for 2 to 6 hours, oxygen in presence of ammonia, chlorine 
at 20 to 40°, bleaching powder, nitric acid, nitrogen trioxide, 
alkalis, and water under pressure, then hydrolyzed with con- 
centrated sulphuric acid. In every case the treatment reduced 
the yield of alcohol. 

According to Classen, 2 the presence of a small amount of an 
acid or salt, differing from the main acid, decidedly increases the 
yield of sugar; thus if sulphuric acid is used, a small amount of 
hydrochloric or sulphurous acid should be added. Metals or 
oxides, particularly of the iron group, may also be used. 

Kressmann 3 used mixtures of chemicals reacting according 
to the following equations: 

1. H 2 S0 4 + NaCl = NaHS0 4 + HC1. 

2. H 2 S0 4 + 2NaCL = Na 2 S0 4 + 2HC1. 

3. 2KC10 3 + H 2 S0 4 + 10HC1 = K 2 S0 4 + 6H 2 + 6C1 2 . 

4. 2KC10 3 + H 2 S0 4 = K 2 S0 4 + 2HCIO3. 

It is improbable that in any case the reactions as given above 
proceeded to completion, but it is of interest that the yield of 
reducing sugars was practically the same (about 20 per cent) 
in each. Ferrous sulphate added to sulphuric acid was without 
effect. Niter cake alone or mixed with sulphuric acid reduced 
the yield of alcohol 20 to 25 per cent. Phosphoric acid gave only 
15.7 gallons of alcohol per ton. Moore 4 proposes to hydrolyze 
wood with phosphoric acid and ferment the sugars without previ- 
ous neutralization. 

The use of manganese and magnesium chlorides in conjunction 
with hydrochloric acid has been protected. 5 Pink 6 claims a 
yield of 62 per cent of sugars by treating wood with alkalis, then 
hydrolyzing with dilute hydrochloric acid in the presence of 
chloracetic acid. Another method involves boiling the wood with 
formaldehyde under pressure previous to hydrolysis with acids. 7 

1 H. Krull, "Versuche liber Verzuckerung von Cellulose," Diss. Danzig 
(1916). 

2 A. Classen, G. P. 351681 (1917); E. P. 142480 (1920). 

3 F. W. Kressmann, I.e., 54. 

4 H. K. Moore, U. S. P. 1323540 (1919). 

5 G. P. 359866 (1917). 

6L. Pink, G. P. 362232 (1917). 
7 L. Pink, G. P. 362233 (1917). 



352 CHEMISTRY OF CELLULOSE AND WOOD 

The effect of a large number of chemicals, including inorganic 
acids, salts, and aromatic sulphonic acids, has been investigated 
by Sherrard and Gauger. 1 The additions usually caused a 
decrease in the yield of alcohol, while in no case was the yield 
sufficiently increased to warrant a change from the regular 
procedure. 

Source of Ethyl Alcohol. — Korner 2 suggested that the yield 
of fermentable sugars was dependent on the cellulose content of 
the raw material. Gallagher and Pearl 3 asserted that the fer- 
mentable sugar was derived equally from the lignin and cellulose, 
the wood being attacked as a chemical entity, lignocellulose. 
The latter view is certainly erroneous. The alcohol is derived 
almost entirely from glucose and mannose, of which sugars 
mannose is the most important. The influence of the mannan 
of conifers was long overlooked. 4 The comparative absence of 
mannan in hardwoods, and the presence of large amounts of 
pentosans, explain the lower yield of alcohol from these woods. 
The differences in yield between species will be noted in the 
table below. 

Western larch (Larix occidentalis) is of special interest, since 
it contains 8 to 17 per cent of galactan. At the time of its 
discovery, difficulty was encountered in fermenting galactose; 
about 40 per cent of alcohol was obtained with Saccharomyces 
carlsbad I. 5 Kressmann 6 obtained 30 per cent of reducing sugars 
on hydrolyzing the wood, but only 38 per cent was fermented 
with S. cervisice, the galactose remaining intact. Sherrard, 7 
using the same yeast, was able to obtain very efficient fermenta- 
tions by keeping the initial acidity 8 below 5° and the tempera- 
ture between 85 and 90°F.; the conditions must be carefully 

1 E. C. Sherrard and W. H. Gauger, J". Ind. Eng. Chem., 15 (1923), 
63, 1164. 

2 T. Korner, I.e. 

3 F. E. Gallagher and I. L. Pearl, Eighth Int. Cong. Appl. Chem., 
N. Y., 13 (1912), 147-149. 

4 A. W. Schorger, J. Ind. Eng. Chem., 9 (1917), 748. 

6 A. W. Schorger and D. F. Smith, J. Ind. Eng. Chem., 8 (1916), 484. 

6 F. W. Kressmann, J. Ind. Eng. Chem., 7 (1915), 920. 

7 E. C. Sherrard, J". Ind. Eng. Chem., 14 (1922), 948; U. S. P. 1454521 
(1923). 

8 In titrating 10 cubic centimeters of sugar solution, 1 cubic centimeter 
of O.liV sodium hydroxide is equal to 1° of acidity. 



SACCHARIFICATION OF CELLULOSE AND WOOD 353 

observed. Autolyzed yeast sufficient to produce a 2 per cent 
solution was used as a nutrient. When the water-soluble portion 
of the wood and the residue were hydrolyzed separately, 40 
gallons of 95 per cent alcohol were obtained per ton. This is 
the largest yield of alcohol that has been obtained from any 



Yields of Alcohol from Various Species of Woods 1 





Total 

reducing 

sugars, 

per cent 


Per cent of total 
reducing sugars 


Alcohol yields 


Species 


Ferment- 
able 


Unfer- 
mentable 


Per cent 
of wood 


Gallons 
190 proof 
per ton 


Conifers 
Western white pine . . 

Red spruce 

Red spruce 

White pine 

Longleaf pine 

Longleaf pine 

Lodgepole pine 

Norway pine 

Western larch 

Western larch 

Western larch 

Western hemlock .... 
Sugar pine 


21.00 
20.48 
22.06 
20.02 
23.06 
23.25 
21.93 
25.63 
29.72 
30.52 
26.21 
21.15 
18.03 
20.23 
23.61 
21.10 
21.13 

20.53 
18.93 
20.74 
21.24 
17.30 
18.38 
18.30 
16.60 
20.42 
18.19 


74.49 
74.16 
72.67 
75.67 
73.32 
72.49 
67.37 
66.88 
37.89 
57.88 
54.69 
77.63 
72.55 
66.49 
71.44 
67.42 
75.16 

46.29 
34.04 

47.22 
22.22 
50.48 
30.40 
38.86 
26.79 
38.81 
32.86 


25.51 
25.84 
27.33 
24.33 
26.68 
27.51 
32.63 
33.12 
62.11 
42.12 
45.31 
22.37 
27.45 
33.51 
28.56 
32.58 
24.84 

53.71 
65.96 
52.78 
77.78 
49.52 
69.59 
61.14 
73.21 
61.19 
67.14 


7.762 
7.565 
7.956 
7.437 
8.282 
8.330 
7.205 
7.745 
4.977 
8.687 
6.934 
7.622 
6.276 
7.115 
8.537 
6.822 
7.934 

4.288 
3.029 
4.661 
1.995 
4.102 
2.675 
3.205 
1.382 
3.658 
2.392 


23.43 
22.84 
24.01 
22.46 
24.90 
25.16 
21.75 
23.38 
15.03 
26.21 
20.93 
23.01 
18 93 


Sugar pine 


21 47 


White spruce 

Douglas fir 


25.78 
20 59 


Douglas fir 


23 95 


Hardwoods 
Birch 


12 95 


Sugar maple 

Silver maple 

Beech 


9.14 

14.07 

6 02 


White oak 


12 38 


Red oak 


8 07 


Sycamore 


9 67 


Slippery elm 

Red gum 


5.98 
11 03 


Cottonwood. . . 


7 21 







i F. W. Kressmann, I.e., 57-58. 



354 



CHEMISTRY OF CELLULOSE AND WOOD 



wood by hydrolysis with dilute acid. The yield fell to 33 
gallons when the wood was hydrolyzed directly. 

Sherrard and Blanco 1 have quantitatively determined the 
sugars formed by pressure hydrolysis of white spruce. Glucose 
and mannose formed two-thirds of the total sugars. Fructose 
was not detected. 

Per Cent 

Mannose 37 . 7 

Glucose 29.3 

Galactose 6.4 

Xylose 13 . 3 

Arabinose 5.4 

Reducing volatile substances 7.9 

Total 100.0 

The change in the composition of the wood, due to hydrolysis, 
was as follows : 



Wood 


Acetic 

acid by 

hydrolysis 


Methoxyl 


Pentosans 


Cellulose 


Lignin 


Ash 


Original 

Hydrolyzed 


1.32 
0.116 


4.75 
3.99 


10.76 
1.79 


58.33 
39.67 


29.19 
29.63 


0.26 
0.06 



It is surprising that the lignin content did not increase on hydrol- 
ysis, since 32 per cent of the cellulose was lost. 

Saccharification with Concentrated Acids. — Braconnot, 2 inl819, 
found that wood could be dissolved in concentrated sulphuric 
acid and largely converted into glucose. Arnould 3 treated 100 
parts of sawdust with 110 parts of concentrated sulphuric acid, 
let stand 12 hours, diluted, boiled, neutralized with chalk, and 
fermented the sugars with yeast. 

Voerkelius 4 treated one part of wood with seven parts of 70 
per cent sulphuric acid and obtained 67 per cent of sugars, of 
which 70 per cent was fermentable. The yield of alcohol was 
24.8 per cent, indicating practically complete saccharification 



1 E. C. Sherrard and G. W. Blanco, J. Ind. Eng. Chem., 15 (1923) 
611-616. 

2 H. Braconnot, Ann. chim. phys., [2] 12 (1819), 172. 

3 J. E. Arnould, Compt. rend., 39 (1854), 807-808. 

4 G. A. Voerkelius, Wochbl. Papierfabr., 42 (1911), 852. 



SACCHARIFICATION OF CELLULOSE AND WOOD 355 



of the hexosans. Similar experiments were made by Hagglund, 1 
who treated the wood with an equal weight of 70 per cent 
sulphuric acid. He obtained the following yields: 







Sugar on 


Alcohol per 


Time, hours 


Temperature 


dry wood, 


long ton, 






per cent 


liters 


24 


18° 


23.5 


80 


48 


18 


54.8 


158 


3 


40 


29.1 


120 


5 


40 


34.9 


125 


3 


50 


26.6 


105 


6 


50 


23.9 


102 


2 


75 


16.0 




4 


75 


14.0 





Attempts have been made to reduce the amount of acid. 
Classen 2 subjected a mixture of 1 part of wood and 0.75 part of 
sulphuric acid of 57°Be. to great pressure, whereby the mass 
became warm; on diluting with 4 parts of water and boiling, 40 
per cent of the wood was converted into sugars. Wohl 3 employs 
powerful agitation and pressure; in this way 1.5 parts of sulphuric 
or hydrochloric acid suffice for 1 part of wood. 

Hydrochloric acid is used in two ways: the wood is treated 
with hydrochloric acid of a concentration of 40 per cent or 
above; 4 or moistened with water and saturated with hydrogen 
chloride gas. The latter procedure was patented in 1880 in 
both England and Germany. 5 Wohl 6 subsequently rediscovered 
the method, but his patent application in Germany had to be 
withdrawn in view of the prior art. Krull 7 moistened wood with 
three parts of water, saturated it with hydrogen chloride, and 

*E. Hagglund, J. prakt. Chem., 91 (1915), 358; "Die Hydrolyse der 
Zellulose und des Holzes," Stuttgart (1915), 52 pp. 

2 A. Classen, G. P. 111868 (1899). 

3 A. Wohl, E. P. 146860 (1920). 

4 R. Willstatter and L. Zechmeister, Ber., 46 (1913), 2401; L. Zech- 
meister, Diss. Zurich (1913). 

6 A. M. Clark, E. P. 268 (1880); E. S. Dauziville, G. P. 11836 (1880). 

6 A. Wohl, W. 39800 (1912). 

7 H. Krull, Diss. Danzig (1916), 51; A. Wohl and H. Krull, Cellulose- 
chemie, 2 (1921), 1-7. 



356 



CHEMISTRY OF CELLULOSE AND WOOD 



completed the hydrolysis in the usual way. He obtained the 
following yields of reducing sugars and alcohol : 



Substance 


Sugar, 
per cent 


Alcohol, 
per cent 


Alcohol in per cent 
of theoretical amount 
obtainable from sugar 


Ground wood pulp 


62.59 
60.95 


15.48 
18.00 


50.0 


Spruce wood 


60 8 







Other methods involve treatment of the wood with an equal 
weight of 40 per cent hydrochloric acid, followed in a short time 
by saturation with hydrogen chloride. The excess acid is 
removed by suction, and saccharification completed by diluting 
with water and boiling. 1 Terrisse and Levy 2 treat the wood 
with two parts of 33 per cent acid and add hydrogen chloride 
until the concentration of the acid reaches 41 per cent. Kocher 3 
claims that almost complete saccharification is obtained with a 
40 per cent acid in 24 to 48 hours at the ordinary temperature, 
and in much less time by working under pressure at a temperature 
of 40 to 60°. 

Vernet 4 describes the saccharification of wood with hydro- 
chloric acid. The apparatus resembles a Herreshoff pyrites 
burner; the wood, treated with aqueous hydrochloric acid, is 
placed in thin layers in a series of trays, and stirred by rotating 
arms while hydrogen chloride gas is introduced. The excess 
acid is removed by heat and suction. All parts of the apparatus 
in contact with the acid are made of "prodorite," a special 
material which is resistant to hydrochloric acid up to a tempera- 
ture of 120°. The yield of absolute alcohol is 250 liters per long 
ton of sawdust, or about 19 per cent by weight. 

The recovery of the large amounts of acids is a serious problem. 
Benesch 5 proposes to use ether for extracting the concentrated 
sulphuric acid, in large part, after the first stage of the reaction. 



1 Chem. Fabrik Rhenania, F. L. Schmidt, and G. A. Voerkelius, 
G. P. 304399 (1917). 

2 H. Terrisse and M. Levy, E. P. 143312 (1920). 
3 R. A. Kocher, E. P. 107219 (1916). 

4 G. Vernet, Chimie & Industrie, Special No. (May, 1923), 654-657. 
*E. Benesch, G. P. 368399 (1920). 



SACCHARIFICATION OF CELLULOSE AND WOOD 357 

Sufficient acid remains to complete the hydrolysis by diluting 
the residue with water and heating. Budnikow and Solotarew 1 
place the dilute sulphuric acid solution of the sugar in the 
cathode compartment of an electrolytic cell; the cathode is 
surrounded by a porous pot. On electrolysis, 91.5 per cent of 
the sulphuric acid is recovered in the anode compartment, with 
the consumption of 7.215 kilowatts per kilogram of acid. The 
power consumed is worth more than the acid recovered. 

The most practical method of recovering hydrochloric acid 
appears to be by the use of a vacuum. According to Budnikow 2 
and Sworykin, 98.74 per cent of the hydrochloric acid can be 
recovered as chlorine by electrolysis, using platinum electrodes. 
The sugar content of the solution surrounding the cathode is 
unchanged. It is claimed that a purer sugar solution is obtained 
in this way than by the use of dilute sulphuric acid, which is 
precipitated with calcium carbonate. An objection to the use 
of hydrochloric acid is the necessity of removing the chlorine 
ions, towards which yeast is intolerant; 3 this cannot be accom- 
plished as economically as with sulphuric acid. 

Products. — The ethyl alcohol obtained from wood is remark- 
ably pure; it contains 0.5 per cent fusel oil, a trace of methyl 
alcohol, and no acetone. 4 

The hydrolysis of coniferous woods gives about 2.0 per cent 
of acetic and formic acids, of which acetic acid constitutes about 
80 per cent. 5 Only about 10 per cent of each acid is obtained in 
relieving the pressure in the digester, so that the recovery would 
not pay. Pringsheim 6 prepared a stock food by treating wood 
with dry steam and gaseous hydrochloric acid (1.5 per cent) 
at a temperature of 120 to 140°. He obtained as by-products 
6.04 per cent of acetic acid, 1.68 per cent of acetone, and 3.5 
per cent of furfural. 

1 P. P. Budnikow and P. W. Solotarew, Chem. Zentr., 93, IV (1922), 
324; P. P. Budnikow, Z. angew. Chem., 36 (1923), 326-328. 

2 P. P. Budnikow and A. I. Sworykin, Z. angew. Chem., 35 (1922), 677. 

3 F. W. Kressmann, I.e., 54. 

« G. Foth, Chem. Ztg., 37 (1913), 1298. 

5 F. W. Kressmann, I.e., 48, 84. 

6 H. Pringsheim, Cellulosechemie, 2 (1921), 123-124. 



358 CHEMISTRY OF CELLULOSE AND WOOD 

Only qualitative tests appear to have been made f or laevulinic 
acid in the hydrolyzate. Bente 1 boiled fir wood with 5 per cent 
sulphuric acid for 8 days and obtained a small amount of laevu- 
linic acid. Noyer 2 has calculated that a long ton of sawdust 
could give 250 kilograms of laevulinic acid, which might serve as 
a raw material for synthetic rubber. The acid, on distillation 
with phosphorous trisulphide, gives mainly a-methylthiophene, 
which on reduction yields isoprene. 

1 F. Bente, Ber., 8 (1875), 416-418. 

2 G. Noyer, Caoutchouc & gutta-percha, 13 (1916), 8804-8805. 



CHAPTER XI 



THE ACTION OF VARIOUS REAGENTS ON WOOD 



Wood is almost inert to water, except at high temperatures, 
and is dissolved with difficulty by all cellulose solvents. Bases 
attack the lignin preferentially, but at high temperatures the 
caustic alkalis convert both cellulose and lignin into aliphatic 
acids. Hot acids disintegrate wood by hydrolysis of the carbo- 
hydrates. Oxidizing agents attack lignin profoundly and con- 
vert the cellulose to oxycellulose. The decomposition of wood 
under atmospheric conditions is due to a slow oxidation. 

Action of Heated Water. — Wood becomes soft and pliable 
when heated with water, and at sufficiently high temperatures 
undergoes hydrolysis. Tauss 1 heated wood with water at various 
pressures up to 20 atmospheres and found that maximum solu- 
bility was reached at 5 atmospheres. Repetition of cooking 
did not greatly increase the extract. A single cook at 10 atmos- 
pheres dissolved 13.68 per cent of birch, and 11.78 per cent of 
spruce; at 20 atmospheres, only 3.34 per cent of birch dissolved. 
None of the sugars formed were identified. Filter paper became 
gelatinous at a pressure of 20 atmospheres and dried to a hard 
mass. Schwalbe and RobinofP have shown that pure cellulose 

Solubility of Wood in Water at 5 Atmospheres' Pressure 







Cook number 


Total, 
per cent 


Wood 


1, 

per cent 


2, 
per cent 


3, 
per cent 


Beech 


f Dry extract 

) Sugar in extract 

\ Dry extract 

\ Sugar in extract 


21.60 
9.85 

15.40 
8.00 


4.00 
1.19 
3.07 
1.00 


1.25 
0.15 
0.71 
0.08 


26.75 


Spruce 


11.19 

19.18 

9.08 



1 H. Tauss, Dingier s polytech. J., 273 (1889), 276-285. 

2 C. G. Schwalbe and M. Robinoff, Z. angew. Chem., 24 (1911), 256-258. 

359 



360 CHEMISTRY OF CELLULOSE AND WOOD 

undergoes very slight change on being heated with water at 
pressures up to 20 atmospheres; a temperature of 150° should 
not be exceeded, however, as the decomposition becomes notice- 
able at this point. 

Wood is converted into a carbonaceous mass resembling coal 
when heated with water at a temperature of 250°. 1 

The sugars produced by pressure hydrolysis were more care- 
fully examined by Koch. 2 He used the wood of aspen (Popu- 
lus tremula) and spruce (Picea excelsa) which had been purified 
by treatment in the cold with water, 5 per cent hydrochloric 
acid, water, 5 per cent ammonium hydroxide, and finally with 
10 per cent sodium hydroxide to remove wood gum. The 
purification resulted in great loss; starting with 1 kilogram, 520 
grams of aspen and 550 grams of spruce were obtained as a 
purified residue. Purified spruce, heated with water at 5 atmos- 
pheres' pressure (150°), gave a syrup containing: 

Pee Cent 

Laevulinic acid 7.8 

Mannose 29 . 2 

Glucose 23 . 9 

Galactose 

Fructose (Seliwanoff reaction) Trace 

No pentoses were detected, but the syrup gave 8.9 per cent 
furfural. The furfural-yielding substances were largely destroyed 
by evaporating the syrup at 100°. The survival of mannose in 
the purified wood is noteworthy. 

Aspen gave mannose, galactose, glucose, and fructose (trace). 
Xylose and arabinose were not found. Celluloses prepared with 
Schulze's reagent were also tested. He concludes that, while 
mannose and galactose are formed from lignocelluloses, they 
originate from the lignin complex and not from the cellulose. 
Glucose is derived from the cellulose. Galactose and mannose 
are formed readily at low temperatures and destroyed at 185°. 
Fructose is formed only in traces, but there is more in spruce 
than in poplar. The presence of sugars other than glucose 
shows that the hemicelluloses had not been completely removed, 
in spite of the purification. 

1 S. Stein, Chem. Zentr., 72, II (1901), 950. 

2 W. Koch, "Verhalten von Lignocellulose beim Erhitzcn mit Wasser 
unter Druck," Diss. Freiburg (1909), 39 pp. 



THE ACTION OF VARIOUS REAGENTS ON WOOD 361 



The amount of sugar formed from a wood gel is somewhat 
lower than from normal wood. 1 In the following table, Norway 
pine was heated with water in an autoclave, held at the maximum 
pressure for 15 minutes, and the pressure released. 





Steam pressure, 
pounds 


Sugars from 


Cook number 


Normal 
wood, per cent 


Gelatinized 
wood, per cent 


1 
2 
3 

4 


75 

5 125 
200 , 
250 


2.93 
6.50 
9.03 

7.53 


1.11 

4.79 
8.55 
6.48 



In the manufacture of mechanical pulp, the longest and 
strongest fibers are prepared by the hot process. According to 
Kirchner, 2 the steam suddenly developed when the wood comes 
in contact with the stone exerts a softening action on the incrus- 
tants holding the fibers together, and they are burst apart. The 
brown color of steamed pulp is probably due to oxidation, since 
in the presence of reducing agents no coloration takes place. 3 

The solubility of lignin in water is very slight, though Potter 4 
thought that both cold and hot water produced delignification 
of the xylem, since the extracts gave the lignin reaction. Spauld- 
ing 5 found that very slight delignification could be produced in 
thin sections of wood heated in an autoclave with water at 120°. 
The lignin of spring wood was more soluble than that of summer 
wood. An annual ring of Pinus palustris, boiled in water and 
stained with phloroglucinol, showed a red coloration of the 
summer wood, while the spring wood retained practically its 
natural color. Zeller, 6 using zinc chloriodine, was unable to 
observe any change in the lignification of autoclaved wood, 
though the aqueous extracts gave a faint pink with phloroglucinol. 

1 A. W. Schorger, J. Ind. Eng. Chem., 15 (1923), 812. 

2 E. Kirchner, Wochbl. Papierfabr., 41 (1910), 1995. 

3 C. G. Schwalbe, G. P. 203230 (1908). 

4 M. C. Potter, Ann. Botany, 18 (1904), 121. 

5 P. Spaulding, Missouri Botan. Gardens, 17 (1906), 41-58. 

6 S. M. Zeller, Ann. Missouri Botan. Gardens, 4 (1917), 93. 



362 



CHEMISTRY OF CELLULOSE AND WOOD 



Wood steamed or boiled for mechanical pulp is stained blue 
by zinc chloriodine. 1 Wood loses 3 to 4 per cent of extractives 
by the boiling process, but only 0.1 to 0.2 per cent by steaming. 
Microscopic examination indicated a loosening of the bond 
between the tracheids of the summer wood. 

Heating wood with water under pressure produces methyl 
alcohol, 2 and acetic and formic acids. Bergstrom 3 heated wood 
with water at a pressure of 6 atmospheres for 2 hours, and obtained 
the following amounts of acids: 



Species 


Acetic acid, 
per cent 


Formic acid, 
per cent 


Ratio of formic 
to acetic acid 


Spruce 


f 1.53 
\ 1.17 

f 1.40 
\ 1.24 

3.13 


0.23 
0.21 

0.22 
0.19 

0.16 


1:6.65 


Pine 


1: 5.57 
1: 6.36 


Birch 


1: 6.53 
1 : 19 . 56 



Steaming coniferous wood at 4 atmospheres' pressure for 8 
hours gave a solution containing sugars, resin, tannin, furfural, 
salts, and organic acids. 4 No furfural is obtained by steaming 

Per cent 

Acetic acid . 213 

Formic acid . 32 

Sugar 5.20 

at atmospheric pressure even in the presence of formic and acetic 
acids. 5 The formation of furfural was first noted by Williams. 6 
He obtained 0.63 per cent of furfural by steaming wood at 100 
pounds' pressure for 4 hours. 

The steaming of 27 solid cubic meters of wood gave: 1 

1 F. A. Zacharias, J. Soc. Chem. Ind., 31 (1912), 582. 

2 H. Bergstrom, Papier-Fabr., 8 (1910), 506, 736. 

3 H. Bergstrom, Papier-Fabr., 11 (1913), 305-306. 

4 E. Heuser, Chem. Ztg., 38 (1914), 126-127. 

B E. Heuser, Z. angew. Chem., 27 (1914), 654-655. 

6 G. Williams, Chem. News, 26 (1872), 231. 

7 C. Franck, Papier-Fabr., 17 (1919), 1019-1020. 



THE ACTION OF VARIOUS REAGENTS ON WOOD 363 

Kilograms 

Amines (principally methylamine) . 20 

Acetone . 14 

Methyl alcohol 5 . 12 

Furfural 8.80 

Methylfurfural (?) 0. 12 

Volatile acids (principally acetic) 40 . 00 

Sugars 80.00 

Humus substances 147 . 00 

Action of Cellulose Solvents. — Wood dissolves reluctantly 
in all cellulose solvents. Soon after the discovery of Schweizer's 
reagent, it was observed that wood was scarcely affected by it. 
The conclusion was accordingly drawn that the lignin must be 
chemically combined with the cellulose. 1 Webster 2 states that 
plant fibers without exception dissolve more or less rapidly 
in contact with metallic copper and strong ammonia; conflicting 
statements are due to the use of cuprammonium solutions of 
different activity. 

Hoffmeister 3 found that by subjecting lignocelluloses to alter- 
nate treatment with hydrochloric or acetic acid, and ammonia, 
that incrustants were so modified that the cellulose could be 
completely dissolved with cuprammonium solution. According 
to Ungar, 4 wood consists of portions rich in cellulose and poor 
in lignin which are dissolved, and portions high in lignin which 
are not. Spruce wood after 13 extractions with cuprammonium 
solution gave a residue of 37 per cent. The methoxyl content 
of the soluble portions varied from 2.19 to 5.80 per cent, while 
the residue contained 9.03 per cent. Roughly, the soluble 
fractions contained 15 to 40 per cent of lignin, and the residue 60 
per cent. Willstatter lignin dissolved to the extent of 3 per cent. 

The difficulty in dissolving the cellulose is only partially due 
to the mechanical interference of the lignin. Birch wood was 
ground with cuprammonium solution in a ball mill for 72 hours, 
and the insoluble residue again treated with cuprammonium 

1 C. Cramer, J. prakt. Chem., 73 (1858), 2; E. Fremy, Compt. rend., 
48 (1859), 277; J. Erdmann, Ann. Suppl., 5 (1867), 224; G. Lange, Z. 
physiol. Chem., 14 (1890), 19. 

2 C. S. Webster, J. Chem. Soc., 43 (1883), 25. 

3 W. Hoffmeister, Landw. Jahrb., 17 (1888), 261; 18 (1889), 773; 
Landw. Vers.-Sta., 60 (1898), 359. 

4 E. Ungar, Diss. Budapest (1916), 84, 90, 102. 



364 



CHEMISTRY OF CELLULOSE AND WOOD 



solution. The residue from the second treatment contained 
34 per cent of lignin, 10 per cent of pentosans, and 56 per cent of 
cellulose. 1 

A solution of zinc chloride in twice its weight of concentrated 
hydrochloric acid is stated to generally dissolve jute and ligno- 
cellulose. 2 Ungar 3 heated spruce with a solution of 100 grams 
of zinc chloride in 55 cubic centimeters of water at 70 to 80°. 
Seven extractions gave an insoluble residue of 32 per cent. The 
reagent first demethylates, then hydrolyzes. 

The solvent action on wood of thiocyanate solutions is very 
slight. 4 

The thiocarbonate reaction, applied to thin wood shavings, 
gave a product having none of the characteristic properties of 
viscose. 5 The wood of Mschynomene aspera swelled and became 
gelatinous, but only 20 to 30 per cent went into solution. 6 Less 
than 10 per cent of elder pith was made soluble. 7 

There is a progressive increase in the solubility of wood by 
the thiocarbonate reaction, beginning with wood flour and ending 
with gelatinized wood. 8 Wood ground in a ball mill with a 
mercerizing solution of sodium hydroxide, then treated with 
carbon bisulphide, showed the following solubilities: 

Per Cent 

Aspen 81.80 

White pine 69.85 

The soluble portion precipitable by acid contained considerable 
lignin : 



Precipitate from 


Lignin, 
per cent 


Pentosan, 
per cent 


Pentosan-free 

cellulose (by 

difference), 

per cent 


Aspen (Populus tremuloides) . . . 
White pine (Pinus strobus) .... 


11.70 

22.20 


10.93 

8.32 


77.37 
69.48 



1 A. W. Schorger, J. Ind. Eng. Chem., 16 (1924), 141. 

2 C. F. Cross and E. J. Bevan, Chem. News, 63 (1891), 66. 

3 E. Ungar, I.e., 96. 

4 H. E. Williams, J. Soc. Chem. Ind., 40 (1921), 221T. 

5 C. F. Cross, E. J. Bevan, and C. Beadle, J. Chem. Soc., 67 (1895), 444. 

6 W. C. Hancock and O. W. Dahl, Chem. News, 72 (1895), 18. 

7 C. F. Cross and E. J. Bevan, "Researches" (1895-1900), p. 137. 

8 A. W. Schorger, unpublished results. 



THE ACTION OF VARIOUS REAGENTS ON WOOD 365 

Wood that has been subjected to hydrolysis surrenders its cellu- 
lose readily and gives a thiocarbonate solution of low viscosity. 
The recovered cellulose contained: 

Per Cent 

Pentosan . 63 

Lignin 1 . 04 

"Cellulose" 98.33 

Action of Bases. — Ammonia, for a weak base, has a high solvent 
action on wood, tannins, coloring matters, resins, lignin, pento- 
sans, etc. being removed. According to Hoffmeister, 1 wood 
extracted with concentrated ammonia for several days no longer 
gives the lignin reaction. The residue from evaporation of the 
solvent smelled strongly of vanillin. Complete extraction of 
wood may take months. 2 

When yellow birch was ground in a ball mill with concentrated 
ammonia, 14.45 per cent of the wood dissolved 3 The solubility 
of wood passing a 100-mesh sieve, when digested in the cold with 
a 2 per cent solution of ammonia, NH 3 , for 24 hours was: 4 

Ammonia 

Extract, 

Species Per Cent 

White pine 12.96 

Sugar maple 8 . 44 

Aspen 8.08 

Yellow birch 7.06 

The alkaline earths have little effect on wood. The use of 
alkalis for removing lignin to obtain wood pulp is treated 
on page 403. 

Schleiden 5 observed that when raspings of Scotch pine were 
treated with excess of a solution of potassium hydroxide and 
heated until a crust formed, the residue stained a bluish black 
with iodine. Long boiling with water destroyed this property. 

Tauss 6 heated cellulose (Swedish filter paper), spruce wood, 
and beech wood with various strengths of sodium hydroxide at 
different pressures. Spruce, after three successive boilings with 
8 per cent sodium hydroxide at atmospheric pressure, lost 49.19 

1 W. Hoffmeister, Landw. Jahrb., 17 (1888), 261. 

2 W. Hoffmeister, Landw. Vers.-Sta., 48 (1897), 409. 

3 A. W. Schorger, J. Ind. Eng. Chem., 16 (1924), 141. 

4 A. W. Schorger, J. Ind. Eng. Chem., 15 (1923), 812. 

5 M. J. Schleiden, Ann., 42 (1842), 300. 

6 H. Tauss, Dinglers polytech. J., 276 (1890), 411-428. 



366 



CHEMISTRY OF CELLULOSE AND WOOD 



per cent of its weight, and beech 54.68 per cent. The losses 
obtained by a single treatment with alkali for 3 hours, are given 
below : 



NaOH 


Pressure, 
atmospheres 


Loss in weight of 


solution, 
per cent 


Cellulose, 
per cent 


Spruce, 
per cent 


Birch, 
per cent 


3 
3 
3 

8 

8 

14 

14 




5 

10 

5 

10 

5 


12.07 
15.36 

20.28 

21.88 
77.33 


28.37 
50.96 
70.31 
56.91 
74.32 
35.45 
97.13 


30.25 
55.66 
65.59 
64.36 
70.66 
46.43 
91.48 



With increase in temperature and strength of alkali, the wood 
approaches complete solubility. 

Formation of Oxalic Acid. — At high temperatures, wood in the 
presence of the caustic alkalis is oxidized to acetic, formic, and 
oxalic acids, the latter predominating. Beyond solution of the 
lignin, wood does not undergo much change at temperatures 
below 180° in the absence of air. Hoppe-Seyler 1 recognized no 
change in filter paper heated with a strong solution of potassium 
hydroxide at 200°. At a temperature of 240°, he obtained cate- 
chol, protocatechuic, oxalic, formic, and acetic acids. The gas 
consisted mainly of hydrogen, with some methane. LangeV 
method for determining cellulose is based on heating wood with 
a strong solution of the caustic alkalis at a temperature of 140 
to 180° for an hour; the incrusting substances pass into solution, 
leaving a residue (51 to 55 per cent) of modified cellulose. 

Gay-Lussac, 3 who discovered the formation of oxalic acid by 
fusing carbohydrates with alkalis, stated that sodium hydroxide 
could be used as well as potassium hydroxide. Possoz 4 found 
that equal parts of potassium and sodium hydroxide could be 
used, but that sodium hydroxide could not be used alone. He 

1 F. Hoppe-Seyler, Z. physiol. Chem., 13 (1889), 77, 79. 

2 G. Lange, Z. physiol. Chem., 14 (1890), 283-288. 

3 J. L. Gay-Lussac, Ann. chim. phys., [2] 41 (1829), 398-402 
4 L. Possoz, Compt. rend. 47 (1858), 207, 648. 



THE ACTION OF VARIOUS REAGENTS ON WOOD 367 



obtained 70 parts of oxalic acid by fusing 100 parts of sawdust 
and 300 parts of potassium hydroxide. Dale 1 reduced the 
method to commercial practice. 

The reaction was studied by Thorn, 2 who found that sodium 
hydroxide could be used alone but that the yield of oxalic acid 
was low. He used a mixture of 50 grams of sawdust and 100 
grams of alkali, in the form of a solution of 42°Be., and heated 
it in layers 1 centimeter thick on an iron plate for 1.5 hours. 
The most economical ratio of KOHiNaOH was 40:60. The 
use of air preheated to 100 to 120°, though it accelerated the 
reaction, and the use of manganese dioxide did not increase 
the yield. Coniferous woods gave 94.7 per cent oxalic acid and 
hardwoods 83.4 to 93.1 per cent. 



Yields of Oxalic Acid by Alkaline 


Fusion of Sawdust 


Ratio 
KOH:NaOH 


Temperature 


Oxalic acid, 
C 2 H 2 04.2H 2 0, per cent 


0: 100 


200-220° 




33.14 


10:90 


230 




58.36 


20:80 


240-250 




74.76 


30:70 


240-250 




76.77 


40:60 


240-250 




80.57 


60:40 


240-250 




80.08 


80:20 


245 




81.24 


100:0 


240-250 




81.23 



Preen 3 fused one part of sawdust with two parts of sodium 
hydroxide and obtained 35.2 per cent of oxalic acid at 200°, 
and 30.0 per cent at 220°. Fusion with four parts of alkali at 
240° increased the yield to 41 per cent. Various methods 
involving technic and the use of catalysts have been protected. 4 

1 J. Dale, E. P. 2767 and 3031 (1856). 

2 W. Thorn, Dinglers polytech. J., 210 (1873), 24-39. 

3 F. H. E. Preen, Chem. Age, 32 (1924), 72. 

4 J. Dale, E. P. 2137 (1872); Capitaine and von Hertling, G. P. 84230 
(1895); G. F. Zacher, E. P. 2308 (1897); H. Plater-Syberg, E. P. 23682 
(1893); Effront, F. P. 373157 (1906); A. Deiss and C. J. Fournier, F. P. 
406722 (1909); C. Ellis, U. S. P. 1001937 (1911); H. O. Chute and K. P. 
McElroy, U. S. P. 1002034 (1911); L. W. Andrews, 1018092 (1912); 
1065577 (1913); F. Mori and J. Maeda, Jap. P. 37100 (1920), C. A., 16 
(1922), 105. 



368 CHEMISTRY OF CELLULOSE AND WOOD 

Droste 1 uses an alkaline permanganate solution, while Hofmann 2 
heats wood with nitrates or chlorates in the presence of 
magnesium salts. 

The alkaline fusion process has been largely supplanted by a 
synthetic method. Carbon monoxide is heated with a solution 
of sodium hydroxide to obtain sodium formate, which on heating 
to a temperature of 350 to 400° decomposes into hydrogen and 
oxalic acid. 

CO + NaOH = H.COONa. 

COONa 
2H.COONa = | + H 2 . 

COONa 

Thorn 3 was of the opinion that the oxalic acid was formed 
almost entirely from the carbohydrates. Von Hedenstrom 4 
obtained 124 per cent of oxalic acid from cotton and 122 per 
cent from oak sawdust, using four parts of potassium hydroxide 
and a temperature of 280°. Preliminary extraction of the wood 
with 10 per cent potassium hydroxide reduced the yield of oxalic 
acid, from which it was concluded that some of the oxalic acid 
was derived from the lignin. Lignin does give oxalic acid, but 
under the usual conditions of fusion none is obtained. 5 

The formation of acetic acid by the fusion of wood with 
potassium hydroxide was observed by Erdmann. 6 The acetic 
acid arises partly by saponification of acetyl groups, but mainly 
by oxidation. Better yields are obtained with potassium 
hydroxide than with sodium hydroxide, and with three parts of 
alkali than with one. 7 The following yields were obtained with 
three parts of potassium hydroxide : 

1 A. Droste, G. P. 199583 (1906). 
2 K. Hofmann, G. P. 277733 (1913). 

3 W. Thorn, I.e. 

4 A. von Hedenstrom, Chem. Ztg., 35 (1911), 853-854. 

5 E. Heuser, Papier-Fabr. Fest u. Auslandsheft, 19 (1921), 75; E. Heuser 
and A. Winsvold, Cellulosechemie, 2 (1921), 113. 

6 J. Erdmann, Ann. Suppl, 5 (1867), 228. 

7 C. F. Cross, E. J. Bevan, and J. F. V. Isaac, J. Soc. Chem. Ind., 11 
(1892), 966-969. 



THE ACTION OF VARIOUS REAGENTS ON WOOD 369 



At 150° 

acetic acid, 

per cent 



At 200 to 250 c 

acetic acid, 

per cent 



Cane sugar 

Hydrocellulose (from cotton) 

Jute 

Pine wood 




33 

29 
37 

28 



The optimum conditions for the formation of formic and 
acetic acids were fully investigated by Mahood and Cable. 1 
At the lower temperatures, about equal parts of formic and acetic 
acids are obtained. At a temperature of 230°, the yield of 
acetic reaches a maximum of 19 per cent; the yield of oxalic acid 
is about 60 per cent. About 2.3 per cent of methyl alcohol is 
formed during the fusion. 

Action of Acids. — The hydrolysis of wood for the produc- 
tion of fermentable sugars is treated in Chapter X. Wood 
impregnated with a dilute acid, such as sulphuric, and allowed 
to dry is converted into brittle hydrocellulose. Charring takes 
place at elevated temperatures. The same effect is obtained 
with salts which hydrolyze readily to form mineral acids. 

Kolbe 2 states that salicylic acid is destroyed by wood; it disap- 
pears completely from a dilute solution (0.2 gram to the liter) 
and cannot be detected in the wood. The phenomenon is 
probably due to adsorption or condensation. 

Wood and cellulose, when treated with concentrated nitric 
acid in the cold, swell and form a gel which Braconnot 3 termed 
xyloidin. Mulder 4 treated wood with fuming nitric acid; after 
4 days, he obtained a gelatinous product which he considered 
to be pure cellulose. Sacc 5 heated spruce wood with nitric acid 
of sp. gr. 1.40 and obtained a gelatinous substance which he 
believed to be pectic acid, but which was in reality oxy cellulose. 6 



1 S. A. Mahood and D. E. Cable, /. Ind. Eng. Chem., 11 (1919), 651-655. 

2 H. Kolbe, J. prakt. Chem., 21 (1880), 443; 22 (1880), 112. 

3 H. Braconnot, Ann. chim. phys., [2] 52 (1833), 293. 

4 G. J. Mulder, J. prakt. Chem., 39 (1846), 150-155. 

5 F. Sacc, Ann. chim. phys., [3] 25 (1849), 218-230. 

6 J. B. Lindsey and B. Tollens, Ann., 267 (1892), 366. 



370 



CHEMISTRY OF CELLULOSE AND WOOD 



a 


rH 







o 


CO 


£ 


4-T 

on 


o 


3 
Td 

£ 


fe 


rt 


o 


(/J 






CO 

P 


w 


PE| 


(J 




cd 


s 


£ 


S3 






O 




+3 




X 


-S3 


| 




-+-< 


>< 


£ 


« 


o 
O 


m 




Q 


o 


O 


Oh 


^ 


a 






o 








z 


+j 


< 


^ 


n 


C/J 


o 







.2 


Tt* W ■* 


t^ rH 


OS CO rJH 




o3 


00 CO o 


CO CO 


00 05 N 




O 


CO CO T}H 


IO H/l 


CO CO CO 


o 











g 


i^ CO 00 


H « O) 


Ci 00 CM 


CO 


Ui CM O 


H N N 


Tf< rH i-i 


CM 


o 










0(0 05 


OOHH 


O O CO 




cp 


b- b- GO 


N 00 05 


CM ^ Tin 




< 




"" 






t o 


Ui CO lO 


"H CO • 


O O O 




o3 


rH lO CM 


CO 05 • 


O O H^ 




o 


CO iO CO 


TjH IO • 


CO N N 


o 










O 


3 


CO I- CO 


H 05 10 


CO CO rH 


o 


sh 


CM !>• H^ 


H IO Tf 


GO CM IO 


CM 


£ 


" 


" 






CM CO CO 


00 b- CO 


T^ Tj< 1> 




<r> 


o co io 


<N iO CO 


tO^Tf 




< 


rH rH rH 


" " 






o 


rH • o 




IO CM O 




c3 


<* ■ GO 




CO 1> GO 




o 


CM • CO 




CO IO 


o 











a 


CO GO id 


M N N 


CO O CO 


1^ 


1> C5 TtH 


CO t^ CO 


■H^ CM IO 




& 












T* ^ tK 


CO rH CM 


H ^ CO 




© 


CO GO rH 


iO 00 N 


CO 00 CM 




^ 


1— 1 






cy 


2 


CO CO CM 


CO CO CM 


CO CO CM 


a 


3 


t-H 


rH 


rH 




o 








H 


rd 


. . < 


- . < 


■ . , 






^ 










e 


, N 








rO 


g 


ss 






e 


£ 








oo 


S- 


a. 






^ 








n 




rfS 






V 


s 


CJ 







V 


s 

S 


Q 

=0 

<5> 


d 










•to 






o 


v — ' 


c3 co 






CD 
-4-3 


"ft 


i2 3 






'M 


cS 


d a 






^ 


s 


5 



C£ 



THE ACTION OF VARIOUS REAGENTS ON WOOD 371 

Cold strong nitric acid or hot dilute acid oxidizes wood rapidly, 
the lignin, especially, being attacked. Nitric acid of sp. gr. 1.35 
to 1.40, when poured upon sawdust, gives off red fumes followed 
by a dense white smoke, then ignition. 1 Acid of sp. gr. 1.268 
to 1.236 produces vapors but no flame. 2 It is not surprising 
that the practice of silver reclaimers of absorbing a solution of 
silver nitrate and nitric acid in sawdust has led to fires. 3 

Nitric acid, 4 and a mixture of nitric and sulphuric acids, 5 have 
been suggested for producing pulp from wood. Cross 6 used a 
10 per cent solution of nitric acid at a temperature of 80°, and 
proposed the recovery of acetic and oxalic acids from the mother 
liquor. Heden 7 claims the formation of picric acid, in addition. 

Lifschutz 8 heated pine wood for 14 to 16 hours, at a tempera- 
ture of 45 to 60°, with a solution containing 32 per cent of 
sulphuric acid and 18 to 20 per cent of nitric acid; after washing 
with sodium hydroxide, about 40 per cent of cellulose was 
obtained. The yield of oxalic acid was 29 to 30 per cent. The 
gases evolved could be oxidized to nitric acid for reuse. 

According to Cross and Bevan, 9 the reaction is not one of 
simple oxidation. The nitrous acid reacts with methylene 
groups in the lignin to give oximes which are subsequently 
oxidized : 

1. X.CH 2 + O.NOH = X.C:NOH + H 2 0. 

2. X.C:NOH -f O.NOH = X.CO + N 2 + H 2 0. 

The formation of considerable quantities of hydrocyanic acid 
strengthens this view. Nitric acid of a concentration of 4 to 7 
per cent, at a temperature of 60°, converts the lignin into an 
acid derivative poor in nitrogen, having the formula C 2 5H 39 N0 2 5 
and forming salts of the formula C^HsiNO^M// Oxidation 
with dilute nitric acid may be prevented by the addition of urea, 

1 L. Archbutt, J. Soc. Chem. Ind., 15 (1896), 84-85. 

2 R. Hass, Chem. Ind., 8 (1885) 173; Analyst, 6 (1881), 231. 

3 W. L. Wedger, J. Ind. Eng. Chem., 11 (1919), 894. 

4 C. H. Barne and C. M. Blondel, Dinglers polytech. J., 164 (1862), 464; 
172 (1864), 238. 

5 J. Lifschutz, E. P. 1824 (1891). 
6 C. F. Cross, E. P. 409 (1894). 

7 J. E. Heden, G. P. 212838 (1908). 

8 J. Lifschutz, Ber., 24 (1891), 1186-1192. 

9 C. F. Cross and E. J. Bevan, Ber., 24 (1891), 1772-1776, 



372 



CHEMISTRY OF CELLULOSE AND WOOD 



which destroys the nitrous acid; the nitric acid then has only a 
hydrolytic action and behaves like sulphuric or hydrochloric acid. 
Baly and Chorley 1 heated beech wood with a 10 per cent 
solution of nitric acid on the water bath for 4 hours. The gas 
evolved had the following composition: 





N 


N 2 


NO 


N0 2 


HCN 


C0 2 


CO 


Per cent 


18.3 


9.4 


9.3 


33.2 


8.3 


17.2 


4.3 



The products of oxidation were : 

Pee Cent 

Fibrous residue (nearly pure cellulose) 48 . 00 

Volatile acids (mainly acetic, and calculated as such) 11 . 80 

Oxalic acid 3 . 84 

Soluble incrusting substances 26 . 16 

Spruce wood gave only 3 to 4 per cent of volatile acids, as it 
represents a different type of lignification. 

Dry wood will take up 10 per cent of its weight of dry hydrogen 
chloride, in comparison with less than 1 per cent for cotton and 
hydrocellulose. 2 

Action of Salts. — Little information is available on the action 
of salts on wood, except in cases where they have been used for 
wood preservation. Some salts appear to be adsorbed in small 
amounts. 3 The assumed chemical combination with wood of a 
salt such as zinc chloride is usually due to the deposition of an 
insoluble basic salt. 

Boucherie 4 claimed that iron acetate exerted a peculiar action 
on wood, not only preserving it against decay, but rendering it 
very hard and dense. He proposed treating wood with hygro- 
scopic salts to preserve flexibility and elasticity. 

It is an ancient observation that when iron nails are driven 
into wood exposed to air and moisture the surrounding wood is 
rapidly destroyed by oxidation. 5 The iron oxide may cause 

1 E. C. Baly and J. C. Chorley, Ber., 28 (1895), 922-927. 
2 E. Ungar, Diss. Budapest (1916), 58. 

3 P. Casparis, C. A., 15 (1921), 1333. Original papers not available. 

4 A. Boucherie, Ann. chim. phys., 274 (1840), 145. 
e K. Micksch, Chem. Zentr., 91, IV (1920), 554. 



THE ACTION OF VARIOUS REAGENTS ON WOOD 373 



catalytic oxidation by atmospheric oxidation, or by hydrogen 
peroxide. 

Konig 1 states that copper sulphate can be readily washed out 
of hardwoods, but that in coniferous woods it appears to com- 
bine with the resin. Copper salts have a highly protective action 
against decay. In a copper mine in Portugal, dating from the 
time of the Romans under Nerva, Payen 2 records the finding of 
a wheel which was in a good state of preservation. The wood 
contained about 1 per cent of copper sulphate and 3 per cent 
of basic iron sulphate. 

Mercuric chloride is adsorbed from its solutions. 3 When wood 
is immersed in a solution of zinc hydrogen fluoride, ZnFl 2 .2HF, 
containing free hydrofluoric acid, the compound is adsorbed 
as such; if the solution is forced into the wood by pressure an 
excess of fluorine over zinc is taken up. 4 In the case of a mixture 
of zinc chloride and sodium fluoride there is adsorption of 
sodium fluoride but not of zinc chloride, as shown by the 
following: 5 

Analysis of Impregnating Solution 



z 


nc 


Fluorine 


Before, 
per cent 


After, 
per cent 


Before, 
per cent 


After, 
per cent 


1.03 
0.91 
0.71 


1.03 
0.89 
0.70 


1.02 
0.75 
0.94 


0.91 
0.61 
0.86 



Fire Retardants. — Wood cannot be prevented from burning, 
but its ignition temperature can be raised by impregnation with 
suitable salts. The protective action of the salt may be due 
to: (1) water of crystallization; (2) formation of a coating by 
fusion; and (3) formation of non-combustible gases. Salts of 

1 R. Konig, Chem. Zentr., 32 (1861), 588. 

2 A. Payen, Compt. rend., 58 (1864), 1033-1035. 

3 R. Nowotny, Z. angew. Chem., 30 (1917), Ref. 136; Oesterr. Chem. Ztg., 
25 (1922), 102-104. 

4 R. Nowotny, Oesterr. Chem. Ztg., 13 (1910), 81-84. 

5 Apparent adsorption is probably due to error in analysis; it is difficult to 
determine fluorine accurately in the presence of chlorine. 



374 



CHEMISTRY OF CELLULOSE AND WOOD 



class 1 are of little value. Dilution of the air with non-combusti- 
ble gases, other than steam, is of more importance. Glowing 
charcoal is extinguished when the surrounding atmosphere 
contains 8 per cent of carbon dioxide and 9 per cent of oxygen. 1 
Glowing wood bursts into flame when the oxygen content of the 
atmosphere is 28 to 30 per cent by volume, and is extinguished 
when placed in one containing 16 volumes of oxygen. 2 

Wood will not ignite spontaneously until it is heated above 
the point of the exothermic reaction, which is about 275°. 
Sound hemlock ignited at 285° and pine at 300 to 330°. Decayed 
wood ignites at much lower temperatures. 3 

The ignition temperature may be taken as the point at which 
sufficient combustible gases are given off to cause the wood to 
burn; this will vary with the time and the temperature applied. 
Prince 4 used the temperature at which the specimens could be 
held for 40 minutes, before ignition by a pilot light took place. 

Temperatures at Which Wood Can Be Ignited after Heating for 

40 Minutes 







Ignition 






Ignition 




Mois- 


tempera- 




Mois- 


tempera- 




ture, 


ture, 




ture, 


ture, 


Species 


per 


degrees 


Species 


per 


degrees 




cent 


Centi- 




cent 


Centi- 






grade 






grade 


Tamarack 


15.0 


180 


Hemlock 





180 


Tamarack 





168 


Noble fir 


18.0 


195 


Basswood 


9.8 


168 


Noble fir 





187 


Basswood 





168 


Western larch 


19.5 


180 


Longleaf pine 


Air dry 


180 


Western larch. . . . 





157 


Longleaf pine 





157 


Sitka spruce 


19.1 


180 


Red oak 


11.2 


157 


Sitka spruce 





157 


Red oak 





157 


Redwood 


19.4 


180 


Hemlock 


18.5 


157 


Redwood 





157 



1 F. C. G. Midler, C. A., 12 (1918), 561. 

2 P. Anema, Chem. Weekblad, 10 (1913), 1056; W. P. Jorissen, Ibid., 
1057. 

3 H. B. Hill and A. M. Comey, Proc. Am. Acad. Arts Sci., 22 (1886), 482. 

4 R. E. Prince, Proc. Am. Wood-Preservers' Assoc, 10 (1914), 158-172; 
Proc. Nat. Fire Protect. Assoc. (1915). 106-149. 



THE ACTION OF VARIOUS REAGENTS ON WOOD 375 

Of the various salts examined as fire retardants by Banfield 
and Peck, 1 ammonium phosphate was the best. Their results 
are given in the accompanying table. 

Ignition Temperatures of Wood Impregnated with Salts 



Chemical 



Strength of 

solution, 

normality 



Ignition 
tempera- 
ture, degrees 
Centigrade 



Impregna- 
tion* 



Zinc chloride (2) 

Magnesium chloride (3) 

Calcium chloride (3) 

Cupric chloride (1) 

Ammonium chloride (2) 

Bismuth chloride (1) 

Ferric chloride (1) 

Sodium chloride (2) 

Ammonium sulphate (3) 

Ammonium phosphate (3) 

Ammonium phosphate (3) 

Borax (3) 

Boric acid (2) 

Aluminum sulphate (2) 

Potassium alum (1) 

Sodium silicate (3) 

Stannic chloride (1) , 

Zinc ammonium phosphate (3) 



0.5 
0.5 
0.5 
0.5 
0.5 
0.3 
0.5 
0.5 
2.0 
2.0 
2.0 
0.29 

0.25 
0.3 
0.5 
0.5 



295 
311 
312 
270 
297 
265 
284 
295 
317 
298 
350 
307 
299 
289 
246 
322 
241 
320 



P 

I 
I 
I 
I 
I 
I 
I 
P 
I 
P 
I 
I 
I 
I 
I 
I 
and 



* P — pressure of 50 to 60 pounds per square inch; I = immersion; V =■ vacuum. 

(1) Fire promoters; wood burns below 290°. 

(2) No effect; wood burns between 290 and 300° 

(3) Fire retarders; wood burns above 300°. 

Resistance of Wood to Chemicals. — Wood is an important 
engineering material, due to its inertness to various chemicals in 
comparison with metals under the same conditions. The woods 
most commonly used are the following: 

1. Cypress (Taxodium distichum), 

2. Longleaf pine (Pinus palustris). 

3. Redwood (Sequoia sempervirens) . 

4. Douglas fir (Pseudotsuga taxifolia). 

5. Sugar maple {Acer saccharum). 



1 W. O. Banfield and W. S. Peck, Can. Chem. Met., 6 (1922), 172-176. 



376 CHEMISTRY OF CELLULOSE AND WOOD 

6. White oak (Quercus alba). 

7. Yellow poplar (Populus deltoides). 

8. Tamarack (Larix laricina). 

9. White pine (Pinus strobus). 

10. Norway pine (Pinus resinosa). 

11. Spruce (Picea). 

The action of various chemicals on the first six species listed 
has been studied by Hauser and Bahlman, 1 whose results are 
given in the accompanying table. The following conclusions 
may be drawn: 

1. Acetic acid. There is only slight action at all concentra- 
tions. 

2. Hydrochloric acid. A 5 per cent solution renders redwood 
brittle, while maple and oak are not affected until the concentra- 
tion reaches 25 per cent. Cypress, fir, and pine withstand a 25 
per cent acid, while all woods are destroyed by the concentrated 
acid. 

3. Sulphuric acid. A concentration of 5.0 per cent has no 
effect on the woods, with the exception of redwood, which is 
made brittle by a 1 per cent solution. Cypress and pine are not 
greatly affected by a 25 per cent solution; stronger acids destroy 
all woods. 

4. Nitric acid. A 5 per cent solution affects redwood; a 10 
per cent, oak and maple, but not cypress, fir, and pine. A hot 
5 per cent or a cold 25 per cent solution destroys all woods. 

5. Sodium hydroxide. Maple, oak, and redwood are seriously 
attacked by a 1 per cent solution, though the other species with- 
stand a 10 per cent solution. Pine alone can be used with a 25 
per cent solution. 

6. Sodium sulphide. A concentration of 20 per cent destroys 
only oak and redwood. 

7. Bleaching powder. Cypress, fir, and pine are most resist- 
ant. 

8. Longleaf pine is the most resistant to chemical reagents in 
general. 

Information obtained from the chemical industries showed 
that for pickling tanks cypress was first choice, followed by 

1 S. J. Hauser and C. Bahlman, Chem. Met. Eng., 28 (1923), 159-163; 
cf. C. S. Robinson, J. Ind. Eng. Chem., 14 (1922), 607. 



THE ACTION OF VARIOUS REAGENTS ON WOOD 377 





93 

a 


o 

W 

o 








1 

1 

mm 


ss 

SS 

VS, VB 
S, Shd 
S, Shd 
S, Shd 

VH, Sh, Di, GR 
VH, GR, W 




! 


1 




! 


! 




5 

I I i i i i pq i i I ffl *l i i i i i i ii i i i i 
1 1 1 1 1 1 .11 1,11 ll 1 l LI l l l l l 

m<n mm mm mm % sisi 
m> mm mm mm £;£ 


GO 

3 

a* 

• S 

'o 
PQ 




o 

W 
O 








> 

mm> 


pq SStftftf ; 

> mmooo 

mmmmj*j-j3& '■ 
mmm»»mmmO \ 




:o : 




0203 • ■ 

O2oo : : 


k Oh « 

o - o 

III. 1 p^ccj 1 1 1 1 1 1 - 1 1 1 1 

i . i n n «i oo i i i i i i gg i i i i 

mm mmpmm -mm'p mm'm'mm r# r - 
mm mm>mmm&mm> mmmmm &<£ 


00 

5 


co 

a 


o 

a 

O 








pq 
i > 

' GOgO 


« 33 

P-<> 0302 

£°"5? rim'mmmm ■ 
mm»m»mmmm \ 




'WW ' 

:o2 02 : 




11 N 


o o 

: : : « «: : m i : i* 1 1 1 1 1 iggi 1 1 1 

mm mm^mmmm . mmm mm 
mm mm>mmmm m mmm mm 


XI 
o 

T-H 


h -a 

OP CU Oi 

G ? ^ 


3 H aS o' 

lOiOO OKJOIO ChiOO'O'OO'OhiiJOiO'OOOiOiOO u 3 t-> K*© "5 O "5 
«iO e« HMO i-H<N HO H(N NHN No3(D^H(NHN 




"3 
c 

s 
o 






c 
c 
c 

w 
c 








C 
a 

& 


c 


!2 


!2 


c 

EC 

2 


i 

cc 


w 

c 


c. 

c 


) 2 


6 
o 



^ cl ft 

Oh 03 



MoOl 



an 


.2 


cd <u cd 


M 







3 <3 £ 


t5 


la o M 


>> 






>>>!« 


-O 






■S'frS 


■d is 


s s $ 


oj-d 


.S.S'S 


ft-a 


o o g 


|o 




Nil: 


o5^G 


^Q 



® ^3 



hQ fl > ft M 
I Ijj 

pq,«P. fe 



0> o 
05 £ 



o 



**1 CO CO ^* ^* tc 

?Ti 1 1-1 

03 02 ojjijcq 



378 



CHEMISTRY OF CELLULOSE AND WOOD 





M 

03 
O 


+3 

O 

w 

o 








i 

i n 
> 

coJ> 


w 

ss 
ss 

VS, VB 
S, CI, Shd 
S, CI, Shd 
S, CI, Shd 
Vh, Sh, Di 
Vh, Sh, Di 
Vh, Sh, Di, GR 
WH, GR 




«P4 
OO 

coco 






i i 
i i 


•i 


Ph 

1 1 1 1 1 1 Ml gStf ... Illl 1 n.1111 
1,11,1 ;„- ' pfOpQQ^fcfc ' ' ' ' 1 3 1 1 1 1 
- P pq PSOPVrfjVPfpfpS gg 

gm Smo ->;>»cjoo ^ ** 

,^02 COOK/KB Jgjzxijzminw ^ ^^ 
^^ cococococo> coco coco coco OO 


T 

Q 
n 


"ft 

3 


o 

« 

'o 

O 








pq 
> 

co co- 
co co> 


>cococo 

CO S?g°S?S?-=3-fl 
cococo>>>>coco 


:rt : 
■o ; 




; !! ; 




: i i : : 
: i i : : 


o 
o 

D 


- Oh 

O 

1 1 1 1 1 1 CO 1 1 GO ;.l 1 1 1 1 J 1 ^ Illl 

i i i i i i -i i P pq_- Ph« i i i i i i i tsj'N iiii 

PQ - - K* ftJ 

.CO CO^COCO* .fHrf5J3^f Scocq 
co> co > co co co O co co co Ococo 


W 

w 

o 

o 

GO 

H 

u 

H 


T3 
O 
O 

is 

CD 


o 

O 








U 

«" 

WP5> 
>» 


-CT3 ffl : 

coco tftf ; 
pqpqpqpq~pq WP . 




:coco : 




"•mm : '• 

•COCO ; ; 
"ODCC ' 

:coco : : 


< 

O 
QQ 

a 


CS3CS3 

i PQnP-iPH ii . . zi 
00 -- - I w« ° 

cocococo'co'co'^V^'^^^S?? 2 ^^ - -co 172 Sco . .co"co"coco 
mmv)wmm>m>»»»»wmmmui Oco coco coco coco 




cd cd <d 

O a 


iflioo omoio BHiooio>oo'OH«3 0'Oioooioioo S 3 cd Scoom 

(MiO eS HNO i-i (N i-HCM i-HCN NhN CN +s to +j g iH (M rH <N 

3 ° S2fe^ 




"e3 
o 

1 

4) 

o 








n 

c 

c 

c 


i 

> 

> 
) 

[ 

! 






C 
a 

IE 


C 

H 


o 

!5 


cr 

i 


c 

1 & 



iz 


5 K 


c 


c 




o 
o 



CD 3 


a 


fl**- 1 


T) 


<d: 








O ^ 


a 


»H O 


>i 


f-T3 




u ^ 


CO 


TJ43 


V 


CO P 


-o 


" CD CD'S 


.as 


S 73 






£ o-c-c 


n 


II 


o^oS 



c 7, 

CD <U cd 



""111 

Hill 

coH°>Q 



^2 CD 03 

03^3 a 

g.-g ft 

cd'-^-D^^? 

.« . fc » M 
^ O CDS-- 

tftrr 

pq iWPhPh 



CD O tJ U 

CD co ^ (hX3 

tH -Q -W >> U-& 

£S O J >S 

^ CO CO P* I CO 

^ 1 1 III 

COCOCOPhCQ 

CO >>S 



THE ACTION OF VARIOUS REAGENTS ON WOOD 379 

white pine and resinous pines, such as longleaf and Norway. 1 
Close-grained, resinous longleaf pine is preferable to cypress 
for cold or warm dilute acids, while cypress is best for hot acids. 
Tanks for hydrochloric acid are given a protective coating of 
pitch or asphalt. Tamarack is most suitable for handling a 
warm, dilute (10 to 12 per cent) solution of hydronuosilicic acid. 
White pine and tamarack are preferred for piping dilute organic 
and inorganic acids. 

Hot solutions of cuprous and cupric chloride carbonize wood 
rapidly. Scotch pine of slow growth was more resistant than 
"pitch pine." Spruce had a low resistance. 2 Wood impreg- 
nated with 1.5 to 4.00 per cent solutions of zinc chloride showed 
a decrease in strength, especially if the temperature was raised 
above normal. 3 

Weathering and Natural Coloration of Wood. — On exposure 
to the weather wood darkens and gradually disintegrates to a 
powder. Certain species turn gray and acquire a silky luster. 
Wiesner 4 investigated these phenomena and reached the follow- 
ing conclusions: Moisture is essential for wood to turn gray; 
direct sunlight promotes the change but is not necessary; a 
brown color is due to transformation of the cell membrane into 
humus; the change of wood into a friable mass is largely limited 
to coniferous woods exposed in a horizontal position to atmos- 
pheric influences; the wood substance gradually changes to 
humus, and expansion and contraction, due to unequal distribu- 
tion of moisture, produces a powder; the powder has a higher 
hygroscopicity and ash content than the original ,wood. 

Attack by fungi excepted, the disintegration of wood exposed 
to the weather is without doubt due to the photochemical forma- 
tion of brittle oxycellulose (see p. 383). The latter is reduced 
to a loose powder by the expansion and contraction of the wood. 

The yellow coloration produced by the aging of lignified 
materials may be destroyed or prevented by the use of hydroxyl- 
amine or its acetate. 5 

1 A. W. Schorger, Chem. Met. Eng., 18 (1918), 528-531. 
2 E. Jensch, Z. angew. Chem., 7 (1894), 153-155. 

3 T. R. C. Wilson and E. Bateman, Proc. Am. Wood-Preservers' Assoc, 
17 (1921), 80-88; H. B. Luther, Ibid., 89-114. 

4 J. Wiesner, Akad. Wiss. Wien, 49, II (1864), 61-92. 

5 C. F. Cross, E. P. 104032 (1916). 



380 CHEMISTRY OF CELLULOSE AND WOOD 

Schramm 1 attributes the gray or green color to particles of 
iron, in dust transported by the wind, which react with the tan- 
nins in the wood. It has recently been shown that the gray 
color is due to the spores of fungi deposited in the surface cells. 2 

Freshly felled wood frequently undergoes pronounced color 
changes which are not always due to fungi. Linden wood 
turns green, the principal coloration taking place in the medul- 
lary rays and wood parenchyma. 3 The coloration is independent 
of temperature and takes place in summer as well as in winter. 
Moisture and oxygen are essential, no color being obtained in 
an atmosphere of ammonia or carbon dioxide. The color, pre- 
sumably due to an iron tannate, is destroyed by oxalic acid and 
sodium hydroxide. 

Von Tubeuf 4 records a special case of a living spruce tree whose 
sapwood, with the exception of the three to four youngest rings, 
was colored blue-black from the roots to the extremest branch. 
The phenomenon was traced to the presence of iron salts in an 
adjacent brook. The heartwood, through its inability to trans- 
port water, was not stained. 

A striking example is the red color produced in fresh alder 
wood (Alnus glutinosa). Dreykorn and Reichardt 5 thought 
that "alder red" (phlobaphene) was formed by the action of an 
enzyme on the tannin. The coloring matter is formed only in 
the living cells; air and moisture are essential, but light is not. 6 
Bases increase the coloration, while acids destroy it. Nothing 
further is known. 

The sapwood of alder (Alnus incana Moench), white birch 
(Betula populifolia) , paper birch (Betula papyrifera) , sugar maple 
(Acer saccharum), and red gum (Liquidambar styraciflua) stain 
various shades during spring and summer. Alder and birch stain 
a reddish yellow or rusty color in a few minutes during hot, 
humid weather. Bailey 7 thought that the color was due to the 

1 W. H. Schramm, Jahresber. Ver. angew. BoL, 4 (1906), 140-163. 

2 M. Mobius, Ber. botan. Ges., 42 (1924), 15-18. 

3 F. W. Neger, Naturwiss. Z. Forst. u. Landw., 8 (1910), 305-313. 
4 C. von Tubeuf, Naturwiss. Z. Forst. u. Landw., 9 (1911), 273-276. 

5 F. Dreykorn and E. Reichardt, Dinglers polytech. J., 195 (1870), 
157-171. 

6 F. W. Neger, Naturwiss. Z. Forst. u. Landw., 9 (1911), 96-105. 
7 1. W. Bailey, Botan. Gaz., 50 (1910), 142-147. 



THE ACTION OF VARIOUS REAGENTS ON WOOD 381 

action of oxidizing enzymes, since no color was obtained if the 
woods were boiled in water. 

Judd 1 investigated the brown stain obtained during the kiln 
drying of green, frozen sugar maple and concluded that enzymes 
played no part. The wood did not discolor in the presence of 
hydrogen peroxide, carbon dioxide, sulphur dioxide or steam in 
the absence of oxygen. Wood treated with hot water, steam, 
and various germicides always stained under the influence of a 
warm, humid atmosphere. The color began at the surface and 
traveled throughout the wood. Staining could be prevented 
by the use of a low temperature (50 to 55°) and low humidity 
(60 per cent). 

Western yellow pine and sugar pine stain during kiln drying, 
unless a low temperature is used. The stain is not on the 
surface, but 1 to 2 millimeters beneath it. In this case, the sub- 
stances producing the stain appear to be carried towards the 
surface of the board by water, and deposited where it escapes 
as steam. 

Artificial Coloration of Wood. — At elevated temperatures, 
wood acquires a permanent brown color, due to incipient 
destructive distillation. 2 Prolonged heating in sealed tubes at 
90 to 100° produces a similar darkening. This may be due to 
tannin, since tannic acid began to decompose after 24 hours' 
heating, and after 264 hours, was converted into a black mass. 3 
Oak wood, heated intermittently in a water oven for several 
weeks, gradually darkened and acquired the brown color of old 
oak. 4 

Woods rich in tannin rapidly turn brown on exposure to fumes 
of ammonia. 5 The color is permanent and extends to a depth 
of 2 to 3 millimeters. Woods poor in tannin may be treated with 
a hot 5 to 10 per cent solution of pyrogallol before "fuming" 
with ammonia. 6 

1 R. C. Judd, J. Ind. Eng. Chem., 7 (1915), 920. 

2 H. F. Weiss, U. S. P. 1366225 (1921); Reissue 15316 (1922). 

3 E. Knecht, J. Sot. Dyers Colourists, 36 (1920), 195. 

4 C. K. Tinkler, Biochem. J., 15 (1921), 484. 

5 W. Kolitsch, Mitt. k.k. techn. Gewerbe-Mus. Wien, 11 (1901), 79-80. 

6 H. WiSLiCENus. Z. angew. Chem., 23 (1910), 1441-1446. 



382 



CHEMISTRY OF CELLULOSE AND WOOD 



In the absence of oxygen, oak acquired a reddish color, which 
changed to the brown of fumed oak on exposure to air. 1 Methyl- 
amine and trimethylamine behaved like ammonia. The 
amount of ammonia taken up by oak during fuming was: 





N 


NHs 




Natural 

oak, 
per cent 


"Fumed" 

oak, 
per cent 


Calculated 
addition, 
per cent 


Found by 

distillation 

with NaOH, 

per cent 


Japanese oak 


0.353 
0.308 


1.75 
2.72 


1.69 
2.90 


1.64 


English oak 


2.23 







When 10 grams of oak shavings were covered with dilute ammonia 
and exposed to 100 cubic centimeters of air, the following volumes 
of oxygen were absorbed in a given time: 



Time, minutes 

English oak, cubic centimeters. . 
Japanese oak, cubic centimeters . 




60 
10 
10 



The absorption of oxygen was complete (21 cubic centimeters) 
on standing overnight. 

The color effects of age may be obtained by imbedding the 
wood in alkaline soil; 2 heating it while imbedded in a mixture of 
lime or calcium carbonate and ammonium salts; 3 and with 
dry ammonia. 4 Kleinstiick 5 proposes to season and color wood 
by subjecting it to aldehydes and ketones in the gaseous state, in 
the presence of ammonia, whereby condensation products are 
formed with the protein and tannins. Oxides of nitrogen produce 
a golden shade. 6 

Dyeing Wood. — Wood seldom dyes uniformly, owing to its 
structure and lack of chemical homogeneity. In most cases the 
resistance to the penetrance of dyes is great, even under pressure. 
Dyestuffs dissolved in water and organic solvents seldom pene- 

1 C. K. Tinkler, Biochem. J., 15 (1921), 477-486. 
2 K. K. Schmidt, E. P., 18558 (1907). 

3 K. K. Schmidt et at., U. S. P. 1012283 (1911); 1057284 (1913); Deutsche 
Werkstatten fur Handwerkskunst, F. P. 425085 (1911). 

4 H. W. de Prez, U. S. P. 1015008 (1912). 

* M. Kleinstuck, U. S. P. 1210491 (1917). 
6 L. Doyen, E. P. 1063 (1913). 



THE ACTION OF VARIOUS REAGENTS ON WOOD 383 

trate wood more than 0.5 to 2 millimeters. 1 Oak does not dye 
satisfactorily by immersion, but pine is penetrated to a depth 
of 5 to 20 millimeters in 8 days. Maple, linden, and poplar 
give uniform colors. Pear wood is used for imitation ebony. 
Beech is one of the most satisfactory woods for dyeing, as only 
slight pressures are required. The greatest penetration is 
obtained by the use of crystalline, water-soluble dyes, but these 
are seldom fast to light. 

The wood of the standing tree may be colored by introducing 
dyestuffs or chemicals at the base and allowing them to be trans- 
ported by the sap. 2 Aniline dyes, the suitability of which must 
be determined by experiment, may be employed. Malachite 
green and methylene blue stain uniformly, while eosin forms 
streaks. The salts of certain bases color wood by reacting with 
the lignin. Aniline hydrochloride colors wood yellow, while 
p-phenylenediamine gives a salmon red. Birch trees are colored 
overnight, even as far as the leaves, by a 1 per cent solution of 
aniline hydrochloride. Tannic acid is one of the few chemicals 
that may be used to produce a definite color tone on subsequent 
treatment with other substances. 

Action of Light. — The deteriorating action of light on mechani- 
cal pulp is mentioned below. The increase in acidity is pro- 
nounced, showing that oxidation has taken place. The products 
resulting from the action of light on cellulose are grouped under 
the indefinite heading of oxycellulose, for want of a better term. 

Harrison 3 has studied the effects on cellulose of the action 
of sunlight, mercury light, and radium emanations. Radium 
emanations are especially destructive. The products formed 
are of the nature of hydroxy-acids, and have a strong reducing 
action. In this way is explained the reduction or fading of 
dyestuffs in cloth . 

The fading action of the sunlight on dyed wood does not 
penetrate to a depth exceeding 0.03 to 0.06 millimeter. 4 

1 F. Moll, Z. angew. Chem., 29 (1916), 405-409; M. de Keghel, Rev. 
gen. chim., 14 (1911), 348-355. 

2 M. Kleinstuck, Z. angew. Chem., 26 (1913), Aufsatz., 239-240; L. S. 
Gardner, U. S. P. 952245 (1910). 

3 W. Harrison, J. Soc. Dyers Colourists, 28 (1912), 225-236; 30 (1914), 
206-211. 

4 W. H. Schramm and A. Jungl, Farber-Zeit., 17 (1906), 333-342. 



384 CHEMISTRY OF CELLULOSE AND WOOD 

Cotton fabric exposed at a temperature of 30 to 35° to the 
light from a Cooper-Hewitt mercury vapor lamp, which gives 
a large amount of ultra-violet light, turned yellow and showed 
all the properties of oxy cellulose. 1 Disintegration is attributed 
to a specific physical action, while the oxidation may be due to 
the formation of ozone. 

It may be mentioned that wood has a pronounced action on 
photographic plates in the dark, and produces its own image 
after exposure for 0.5 to 18.0 hours at temperatures up to 55°. 
Charring of the wood on one side causes it to become highly 
active. 2 The active substance appears to occur in the layer 
between the charcoal and the wood. 3 Activity was exercised 
through paper, but not through sheets of glass and metal. 

Russell, 4 found that all woods were active, especially the coni- 
fers; this may be explained by the presence of resins which have 
the same action. 5 Tannins, essential oils, and other organic 
compounds are effective. 6 Oak, beech, Spanish chestnut, and 
sycamore were very active, but ash, elm, and horse chestnut 
only slightly so. The spring wood of Scotch pine affects the 
plate, but the autumn wood does not, unless it is powdered; in the 
larch, the condition is reversed. Oak wood 100 years old was 
still active, as well as the central portion of a piece of oak taken 
from a Dartmoor bog. The activity of all woods could be 
greatly increased by exposing them to sunlight. Blue light was 
as effective as white, but red and green were not stimulating. 

The exact nature of the phenomenon has not been ascertained. 
It is not due to radioactivity, since a sheet of glass or mica 0.001 
inch thick entirely prevents action. Russell is of the opinion 
that the active agent is hydrogen peroxide, since it will pass 
through porous bodies and gelatin. 

Action of Oxidizing Agents. — Reagents such as hydrogen 
peroxide, 7 nitric acid, potassium chlorate, a mixture of nitric 

1 C. Doree and J. W. W. Dyer, /. Soc. Dyers Colourists, 33 (1917), 17-19. 

2 W. J. Russell, Proc. Roy. Soc, 61 (1897), 424-433. 

3 D. Isitani, Physik. Z., 10 (1909), 1003-1005. 

4 W. J. Russell, Proc. Roy. Soc, 74 (1904), 131-134; Trans. Roy. Soc, 
197B (1905), 281-289. 

6 E. von Aubel, Compt. rend., 138 (1904), 961-963. 

6 H. M. Ward, Proc Cambridge Phil. Soc, 13 (1904), 3-11. 

7 C. Marggraf, Kunststoffe, 7 (1917), 165. 



THE ACTION OF VARIOUS REAGENTS ON WOOD 385 

acid and potassium chlorate, etc. have been used for bleaching 
wood. The light color obtained is seldom permanent. 

When potassium persulphate was run into a dilute potassium 
hydroxide solution in which pine wood was being digested at a 
temperature of 60 to 65°, the odor evolved was almost unendur- 
able. 1 A potassium hydroxide solution containing wood gave 
an odor of vanillin or coumarin on electrolysis. 

The action of ozonized oxygen, containing about 1 per cent of 
ozone, on beech wood shavings has been investigated by Doree 
and Cunningham. 2 Unlike jute, the wood was not bleached. 
Ozone attacked the wood rapidly, rendering it friable and pro- 
ducing a nauseous odor through the formation of organic acids 
which were partly volatile in steam. Ozone had practically no 
action on dry wood, but in the presence of moisture there was 
at first a rapid formation of organic acids followed by a drop 
to one-tenth of the initial value. Carbon dioxide was produced 
continuously, there being little decrease in rate after 6 hours. 
The volatile acids, consisting of acetic, formic, and another acid 
having reducing properties, amounted to 2.6 per cent, calculated 
as acetic acid. The non-volatile acid liquor had an acidity 
three to five times greater than the value for volatile acids. 
Oxalic acid and phenols were absent. The solution contained 
reducing substances of an aldehydic or ketonic character. Ozoni- 
zation for 36 hours produced a loss in weight of 40.8 per cent. 
The residue had a high copper number; it gave a large amount 
of furfural, but contained only a small amount of methoxyl. 
The absence of vanillin indicates that lignin does not contain a 
coniferyl alcohol group as postulated by Klason. It is claimed 
that wood may be seasoned by treatment with ozonized air, 3 but 
it is not apparent wherein benefit would accrue. 

Smith 4 has studied the deterioration of mechanical pulp under 
the influence of sunlight, ozone, and temperature. The paper 
was placed between glass plates, thus excluding most of the 
ultra-violet rays, and exposed to the sun for 75 hours. The 

X L. Meyer, Chem. Ztg., 31 (1907), 902. 

2 C. Doree and M. Cunningham, /. Chem. Soc, 103 (1913), 677-686. 
3 M. P. Otto, E. P. 133263 (1919); G. F. Lyon, E. P. 182504 (1921). 
4 W. H. Smith, "Studies on Paper Pulps," Bur. Standards, Tech. Paper, 
88 (1917), p. 11. 



386 



CHEMISTRY OF CELLULOSE AND WOOD 



similarity of the results indicates that the deterioration was due 
in all cases to oxidation. 



Effect of Deteriobation on 


Mechanical Pulp 






Original 
pulp 


Sunlight 


Ozonized 
air 


100° for 
20 hours 


150° for 
20 hours 


Methoxyl (per cent) . 
Furfural (per cent) . . 
Copper number (cor- 
rected) 

Acid number 


5.10 
6.23 

2.0 
16.1 


4.60 
5.41 

7.6 
51.0 


4.74 

5.82 

10.9 
63.2 


4.75 
5.72 

3.00 
27.0 


4.48 
4.64 

11.9 
63.0 



An interesting case of the formation of oxalic acid has been 
called to the writer's attention. A locust insulator pin support- 
ing a high-tension wire (22,000 volts) on the Boston and Maine 
Railroad was converted to oxalic acid to a considerable extent. 
Since ozone does not appear to produce oxalic acid in quantity, 
the active agent must have been nitric acid produced from 
atmospheric nitrogen. 

Wood is capable of yielding small quantities of iodoform. 
Schorger 1 obtained iodoform from the solution resulting from 
grinding aspen in dilute alkali, and suggested that the parent 
substance might be acetone. Huebner and Sinha 2 treated 200 
grams of poplar wood with 4 per cent of its weight of iodine in 
the form of iodine-potassium iodide solution, let stand several 
hours, added dilute sodium hydroxide and steam distilled; 0.25 
gram of iodoform was obtained. Since cotton and purified 
wood cellulose continued to give iodoform on re-treatment, it is 
evident that some of the iodoform originates from the cellulose 
molecule. 

1 A. W. Schorger, J. Ind. Eng. Chem., 16 (1923), 812. 

2 J. Huebner and J. N. Sinha, J. Soc. Chem. Ind., 42 (1923), 258T. 



J 



CHAPTER XII 
PULP PROCESSES AND WOOD PULPS 



Wood, for purposes of illustration, may be likened to honey- 
comb, on the wall (lignin) of which thin layers of a mixture of 
cellulose and lignin have been deposited to form the fibers. The 
framework in wood, known in a broad sense as the middle lamella, 
consists essentially of lignin admixed with hemicelluloses. Proc- 
esses for producing pulp are based on the use of chemicals which 
will dissolve the strongly adhering middle lamella and leave the 
cellulose fibers in the form of loose bundles. Microscopic 
examination of thin sections of wood which have been subjected 
to the action of various chemicals should give valuable informa- 
tion on structure and the course of disintegration. A beginning 
in this direction has been made. 1 

Sulphite Process. — The cooking liquor in the sulphite process 
contains sulphurous acid as the active constituent, though bisul- 
phites are present also. The sulphite liquor is prepared by 
burning pyrites or sulphur. The gas containing the sulphur 
dioxide is washed with water at a temperature of 50° to 75° to 
remove dust and sulphur trioxide, and passed up a tower contain- 
ing limestone or dolomite over which water trickles. In this way 
a solution of calcium and magnesium bisulphites is obtained. In 
comparison with the calcium salt, magnesium bisulphite is 
more soluble and stable, and in case of the formation of sulphuric 
acid, gives the readily soluble magnesium sulphate. Pulps 
prepared with magnesium bisulphite resemble those obtained 
with sodium sulphite. The tower liquor is fortified with sulphur 
dioxide to the point desired before being used in the digester. 

There are three modifications of the sulphite process, the 
Mitscherlich, Ritter-Kellner, and "quick cook." The liquor for 
the Mitscherlich process contains 2.8 to 3.8 per cent total sulphur 

1 A. Abrams, J. Ind. Eng. Chem., 13 (1921), 786; G. J. Ritter, Ibid., 17 
(1925), 1194-1197. 

387 



388 CHEMISTRY OF CELLULOSE AND WOOD 

dioxide, of which 1.7 to 2.7 per cent is free and 1.1 per cent com- 
bined. Indirect heating by steam coils is employed, the steam 
pressure being 85 to 110 pounds. The temperature within the 
digester reaches 105° in 5 hours, 130° in 12 hours, and 135° 
in 16 to 24 hours, at which time the cook is completed. The 
pressure within the digester is 45 to 60 pounds. In the Ritter- 
Kellner process, the liquor contains 4 per cent total sulphur 
dioxide, of which 2.7 per cent is free. Live steam is introduced 
into the digester, where a temperature of 100° is reached after 3 
hours. The maximum temperature of 140 to 150° is reached 
in 8 hours and the cook completed in 15 hours. The pressure 
within the digester is 60 to 80 pounds. The "quick cook" is 
completed in about 8 hours. The liquor contains 5.0 to 5.5 
per cent total sulphur dioxide, of which 1.2 per cent is combined. 
The internal pressure is about 90 pounds and a maximum tem- 
perature of 155° is reached. 

Ordinarily, 30 to 40 per cent of the sulphur dioxide charged 
into the digester is recovered during release of pressure. In 
European practice, the sulphur actually consumed is 110 to 
120 pounds per ton of wood, while in America, where the " quick 
cook" is mainly used, consumption runs from 120 to 180 pounds. 
The yield of pulp varies from 40 to 50 per cent. 

Uniform penetration of the chips before heating contributes 
greatly to the success of the cook. The swelling of wood varies 
directly as the quantity of sulphurous acid in the liquor. 1 Wood 
will in most cases absorb about 250 parts of liquor, though ex- 
treme variations of 150 to 500 parts may be obtained. The wood 
should be neither too dry nor too wet; if too dry, it will absorb 
very little liquor. Impregnation is facilitated by the use of 
" wetting" compounds, such as Turkey red oil and other sul- 
phonated oils which reduce the surface tension of aqueous 
solutions. If wood is impregnated with bisulphite solution, 
preferably by hydraulic pressure, at temperatures between 50 
and 100°, 70 to 80 per cent of the bisulphite is adsorbed and 
retained by the wood during subsequent heating. 2 After 
thorough impregnation, the temperature may be raised rapidly 
to 140° and the digestion completed in 1.5 to 2.0 hours. 

1 C. G. Schwalbe, Wochbl. Papier-Fabr., 54 (1923), 22-23. 

2 C. G. Schwalbe, Papierfabr., 21 (1923), 493-495. 



PULP PROCESSES AND WOOD PULPS 389 

Klason, 1 in 1893, advanced the hypothesis that the action of 
sulphite liquor consisted in the union of sulphurous acid with 
double bonds and aldehyde groups in the lignin, with the forma- 
tion of sulphonic acids. It had been shown by Lindsey and 
Tollens 2 that in spent sulphite liquor the sulphur was present as 
the sulphonic acid. The formation of suphurous acid by dilution, 
and also after titration of the free acid present with iodine, indi- 
cates, in addition, a typical aldehyde addition compound. 3 

Complete extraction of the lignin from 100 grams of spruce 
wood should theoretically require 7.91 grams of sulphurous acid 
and 3.46 grams of calcium oxide. 4 In commercial operations, 
due to incomplete saturation of the aldehyde group in a-lignin, 
only 7.6 grams of sulphur dioxide and 3.2 grams of calcium oxide 
are consumed. Only a-lignin (acrolein-lignin) is capable of 
combining loosely with sulphur dioxide. Of the total 30 per 
cent of lignin in spruce, two-thirds consists of a-lignin and one- 
third of jS-lignin. It has been found that in cooking 100 grams 
of wood the loosely combined sulphur dioxide is 3.5 grams. 5 
On the basis of one molecule of sulphur dioxide to one molecule 
of a-lignin (C22H22O7 = mol. wt. 398), there is present in the 
wood 21.7 per cent of a-lignin. This is in fair agreement with 
Klason's figure of 19.0 per cent. 

Cross and Engelstad 6 consider the mechanism of the sulphite 
process as mainly one of direct sulphonation. Wood can be 
satisfactorily pulped with sulphurous acid alone, using a con- 
centration of 7 per cent and a temperature of 100 to 115°, over 
a period of 15 to 48 hours. Sulphuric acid should be absent. 
"The secondary effect differentiating the by-products of the 
sulphurous acid and bisulphite processes are those of the degree 
of sulphonation, deoxidation of the CO group, and condensation 
of the sulphonated complex." The waste liquors from the 

1 P. Klason, Teknisk Tids., (1893), 17; cf. Svensk Kern. Tid., 9 (1897), 
133; Chem. Ztg., 21 (1897), 261. 

2 J. B. Lindsey and B. Tollens, Ann., 267 (1892), 341. 

3 W. Kerp and P. Wohler, Arb. kais. Gesundh., 32 (1909), 120-143. 

4 P. Klason, Ber., 53 (1920), 1869. 

5 E. Hagglund and C. B. Bjorkmann., Svensk Kern. Tid., 36 (1924), 137. 

6 C. F. Cross and A. Engelstad, J. Soc. Chem. Ind., 43 (1924), 253-257T; 
E. P. 12943 (1922). 



390 CHEMISTRY OF CELLULOSE AND WOOD 

sulphurous acid method differ from those obtained by the calcium 
bisulphite method in their capacity to form gels and in their 
tanning properties. 1 

The secondary reactions taking place in the sulphite process 
are exceedingly complex and their nature entirely unknown. 
It is possible to obtain twice the actual consumption of sulphur 
dioxide in one case as in another; yet the pulps obtained are 
similar in yield and properties. According to Fuchs, 2 there is 
no satisfactory theory of the sulphite process, and in the main 
this is correct. Klason's hypotheses, however, form the best 
working basis available. 

While the main reaction may be sulphonation, Miller and 
Swanson 3 consider hydrolysis important. Cooks were made 
at a temperature of 145°, with a total sulphurous acid concentra- 
tion of 3.80 per cent, of which 2.40 per cent was free and 1.40 
per cent combined. Removal of the lignin practically ceased 
at the end of the eleventh hour; yet the pulps were not of good 
quality. This was not due to cessation of sulphonation, since 
the organically combined sulphur continued to increase with time. 
With a liquor containing 5.60 per cent total sulphur dioxide, of 
which 4.46 per cent was free, the conditions were favorable for 
hydrolysis and good pulps were obtained. In fact, the opinion 
is expressed that the chief reaction in the sulphite process is 
hydrolytic. 4 The factor having the greatest influence on the 
yield and character of the pulp is the free sulphur dioxide present. 
While a certain amount of combined sulphur dioxide is essential 
to a successful cook, the combination of the base and sulphur 
dioxide with the lignin are phenomena of secondary importance. 

Much botanical and chemical evidence indicates that there is 
no chemical union between the lignin and the cellulose. In a 
sulphite cook, simultaneously with sulphonation of the lignin, 
hydrolysis of hemicelluloses and a portion of the cellulose takes 
place, but this may be of minor importance. Some of the 
hemicelluloses appear to be intimately associated with the 

1 C. F. Cross and A. Engelstad, E. P. 202016 (1922). 

2 W. Fuchs, Ber., 54 (1921), 484-490. 

3 R. N. Miller and W. H. Swanson, Paper Trade J., 74, 15 (1922), 295- 
305. 

4 R. N. Miller and W. H. Swanson, Paper Trade J., 77, 14 (1923), 51-52. 



PULP PROCESSES AND WOOD PULPS 



391 



lignin. Whether the successful removal of the lignin depends 
on the hydrolysis of this portion of the hemicelluloses, is unknown. 

Temperature has greater influence on the progress of the pulp- 
ing reaction than the concentration of the acid. 1 Changes in 
temperature that are commercially feasible are of greater effect 
than any changes in the acid concentrations within the usual 
range of technical practice. In the ordinary cook the relation 
between pressure, temperature, and acid concentration are such 
that the beneficial effect of increase in acid concentration are 
largely offset by pressure and temperature limitations, so that 
temperature is left as the dominant factor in controlling the 
progress of the reaction. The lowest average temperature that 
will give the requisite purification, in the necessary length of 
cook, will produce the best quality of pulp. 

In obtaining 2 the experimental data given below, the cooking 
liquor contained 5.50 per cent of sulphur dioxide, of which 4.30 





Yield and Composition of 


Pulps 




Cook 


Pressure 

per square 

inch, 

pounds 


Yield of 
screened 

pulp, 
per cent 


Lignin 
in pulp, 
per cent 


Cellulose 
in pulp, 
per cent 


Yield of cellu- 
lose based on 
wood, 
per cent 


3050 
3051 
3056 
3057 
3071 


65 
70 
75 
80 
100 


48.3 

48.4 
48.2 
48.2 

47.7 


1.7 
2.5 
1.1 
1.2 
1.5 


94.6 
96.4 
96.1 
95.9 
95.3 


45.7 
46.7 
46.5 
46.2 
45.5 



per cent was free and 1.20 per cent combined. The temperature 
was brought to 110° in 1.5 hours and uniformly raised to 148° in 
8.5 hours. It will be noted that the results are very uniform in 
spite of the variation in pressure, hence concentration of acid. 
Contrary to the general opinion, the maximum removal of 
lignin in a quick cook does not occur towards the end of the cook. 3 
Removal of lignin takes place at considerable speed at tempera- 
tures slightly above 100° and proceeds rapidly from this 
point to the end of the cook, which under the conditions obtain- 



1 R. N. Miller and W. H. Swanson, Paper Trade J., 78, 15 (1924), 178- 



181. 



2 R. N. Miller and W. H. Swanson, Paper Trade J., 79, 16 (1924), 48. 

3 R. N. Miller and W. H. Swanson, Paper Trade J., 74, 15 (1922), 298. 



392 



CHEMISTRY OF CELLULOSE AND WOOD 



ing in the particular experiments was at the end of the eleventh 
hour. The loosely combined sulphur dioxide reached a maxi- 
mum at this point and then decreased. It was found that the 
point of maximum production of loosely combined sulphur 
dioxide was an excellent indication of the end of the cook. 
This point does not show that a good pulp will be obtained, but 
does indicate clearly the uselessness of continued cooking. 

Decrease in combined sulphur dioxide may take place at too 
high temperatures or pressures. Seidel 1 assumes that wood 
reacts with sulphurous acid as follows : 

X.CH:CH.Y + H 2 S0 3 -> X.CH 2 .CH.S0 3 H.Y. 

At higher pressures decomposition takes place in accordance 
with the equation: 

X.CH 2 .CH.S0 3 H.Y -> X.CH 2 .CHOH.Y + S0 2 . 

There is thus formed a compound with a greater hydroxyl con- 
tent than that of the original lignin. 

In cooking by the Mitscherlich process, using indirect heating, 
all conditions being identical except time, a rapid drop in lignin 
does not occur until after the ninth hour. 2 At the end of the 

Mitscherlich Sulphite Cooks 
(Maximum temperature, 145°; total S0 2 , 3.8 per cent; combined S0 2 , 1.4 

per cent) 



6 

M 

o 
o 
Q 


a 

o 

|3 


3 

O <D 


Solubility 
in 1 per 
cent NaOH, 
per cent 


'E >-< 
.9 & 


8 "3 

o V 


11 


02 

o 
1c 

-33 g, 


Wood 






15.1 


28.3 


58.4 






630 


7 


80.8 


14.8 


20.9 


50.2 


7.4 


8.2 


629 


9 


62.8 


11.2 


12.0 


45.9 


16.3 


12.5 


626 


10 


64.0 


8.4 


10.6 


49.2 


17.7 


9.2 


625 


11 


51.0 


7.3 


3.9 


43.7 


24.4 


14.7 


634 


11.5 


50.6 


5.7 


4.6 


44.2 


23.7 


14.2 


633 


12 


51.2 


5.6 


3.8 


45.1 


24.5 


13.3 


631 


12.5 


51.6 


5.9 


2.5 


46.7 


25.8 


11.7 


632 


13 . 25 


47.1 


5.3 


3.0 


42 ..7 


25.3 


16.7 


636 


15 


48.4 


5.3 


0.9 


43.1 


27.4 


15.3 



1 H. Seidel, J. Soc. Chem. Ind., 19 (1900), 1033. 

2 M. W. Bray and T. M. Andrews, Paper Trade J., 76, 3 (1923), 49-52. 



PULP PROCESSES AND WOOD PULPS 393 

seventh hour losses of lignin and of cellulose are about equal, 
but from this point on, the lignin disappears faster than the 
cellulose. The portion of the wood soluble in 1 per cent sodium 
hydroxide shows little decrease up to the ninth hour. Removal 
of lignin is always accompanied by loss of cellulose, and it does 
not appear possible to obtain more than 45 per cent of cellulose 
by present commercial methods of cooking. According to 
Sieber, 1 the loss based on total cellulose is of the order of 11 to 
12 per cent. 

The amount of combined sulphurous acid present in the cook- 
ing liquor is of great importance. Cooks made with a total 
sulphur dioxide content of 5.0 per cent, of which 0.30 to 2.09 
per cent was combined with lime, showed that the combined 
sulphur dioxide must not be less than 1 per cent; if lower, the 
screenings, color, and bleach consumption of the pulp are 
increased. 2 Beyond 1 per cent combined sulphur dioxide, these 
factors remain constant. In the case of incompleted cooks, 
the spent liquor may contain only half of the base originally 
present. 3 The remainder is largely present in the pulp, appar- 
ently organically combined, in a form slightly soluble or slowly 
diffusible. In the first stage of the cook the ash content rises 
from 0.48 to 1.8 per cent. 4 This localization appears to be 
independent of the acid content of the liquor, but is largely 
influenced by the moisture content of the wood, and varies 
inversely to it. All other conditions being equal, the wetter 
wood gives a higher yield and a better pulp, though a longer 
cooking period is required. 

The consumption of sulphur dioxide may vary greatly. With 
a low content of calcium oxide, e.g., 3 grams per 100 grams of 
wood, the wood can be completely pulped with the consumption 
of 9 grams of suphur dioxide. 3 This is the lowest amount that 
should be used in practice. The sulphur dioxide consumption 

1 R. Sieber, Papier-Fabr., 22 (1924), 243. 

2 S. E. Lunak, "Effect of Varying Certain Cooking Conditions in the 
Production of Sulphite Pulp from Spruce," U. S. Dept. Agr. Bull., 620 
(1918). 

3 R. N. Miller and W. H. Swanson, Paper Trade J., 76, 9 (1923), 49-51. 

4 C. G. Schwalbe, Zellstoffu. Papier, 1 (1921), 11. 

5 E. Hagglund and C. B. Bjorkmann, Svensk Kem. Tid., 36 (1924), 
133-135. 



394 CHEMISTRY OF CELLULOSE AND WOOD 

increases with the amount of calcium oxide present up to 4.8 
grams, after which there is no essential difference between the 
limits 4.8 to 7.2 grams of calcium oxide. In general, the higher 
the sulphur dioxide content of the liquor, the greater is the con- 
sumption of sulphur dioxide. The loosely combined sulphur 
dioxide is 3.5 grams per 100 grams of wood. The yield of pulp 
calculated on the calcium oxide content of the cooking liquor is 
greatly dependent on the uniformity of the wood. If the lime 
content is too high, a large excess of sulphur dioxide is required 
to bring the cook to completion in the usual time. 

The chemical and mechanical properties of the pulp depend 
on the composition of the liquor and the manner of heating, but 
especially on whether available sulphite (calcium bisulphite) is 
present at the end of the cook. 1 Immediately or sometime after 
disappearance of the available sulphite, not only the pulp but 
the composition of the liquor changes greatly. With a relatively 
high initial lime content and moderate amount of sulphur dioxide, 
a pulp of the greatest strength is obtained if the cook is stopped 
shortly after disappearance of the available sulphite. With 
increasing concentration of sulphur dioxide, the proper time for 
stopping the cook is thrown nearer the point of vanishment of the 
available sulphite. 

Within wide limits of composition of the cooking liquor, with 
respect to calcium oxide and sulphur dioxide, the same yield of 
pulp, namely, 48 to 49 per cent, is obtained if the cook is stopped 
at the proper time. The copper number of the pulp increases 
with the amount of lime present. After disappearance of the 
available sulphite, the Tingle bromine number of the pulps 
increases slowly and then suddenly drops at a certain moment. 
At this point, the pulps show greatly decreased strength. The 
amount of sulphur dioxide firmly bound, and hence to be con- 
sidered as a measure of the sulphur consumption, increases 
greatly as the concentration of sulphur dioxide increases. An 
acceleration of this combination of sulphur dioxide sets in after 
disappearance of the available sulphite. 

According to Miller and Swanson, 2 the reactions involving the 
combination of the base and the sulphur dioxide with the organic 

1 E. Hagglund, Svensk Kern. Tid., 36 (1924), 284-294. 

2 R. N. Miller and W. H. Swanson, Paper Trade J., 76, 10 (1923), 51-54. 



PULP PROCESSES AND WOOD PULPS 395 

matter are more or less independent of the reactions governing 
the yield and quality of pulp, and no rule can be laid down for 
them. Comparable pulps can be obtained with a sulphur 
consumption varying from 262 to 455 pounds of sulphur per ton 
of dry pulp. The number of pounds of lignin removed per pound 
of sulphur in organic combination varied from 5.44 to 6.55. The 
data on loosely combined sulphur contributed little to a knowl- 
edge of the reactions going on within the digester. The values 
for loosely combined sulphur, while approaching a constant, 
have no apparent bearing on the quality of the pulp, and probably 
represent reactions independent of the main pulping reactions. 

It has been found that wood which has been boiled with very 
dilute mineral acid resists resolution by the sulphite process. 
The stronger the acid, the more resistant becomes the wood. 
The same effect is produced if pine wood is given a preliminary 
boiling with 2 to 8 per cent of sodium hydroxide. 1 Coniferous 
wood, in particular, gives a poor cook with sodium sulphite if free 
sodium hydroxide is present within certain limits. 2 Apparently, 
the lignin is resinified by the action of acids and alkalis on the 
aldehyde (?) group and double bonds, so that it has lessened 
capacity to react with sulphite liquor. 

This peculiarity of lignin, apparently, has a direct connection 
with irregularities in the cooking process, such as the formation 
of dark liquors and pulps. Klason 3 in his early work calculated 
that 1 kilogram of wood is capable of combining with 105.6 grams 
of sulphur dioxide and 45 grams of calcium oxide. In practice, 
only about 75 per cent of these amounts are used, since half of 
the theoretical amount of sulphur dioxide would form unstable 
sulphonic acid groups. In order, therefore, to obtain a successful 
cook, it is necessary to have at least 22.5 grams of calcium oxide 
present to combine with the stable ligninsulphonic acids, which 
are strong acids. 

A cook was made, using 131 grams of total sulphur dioxide 
and 45.2 grams of calcium oxide per kilogram of wood. The 
liquor contained a total of 3.40 per cent of sulphur dioxide, of 



1 R. Michel-Jaffard, Chimie & Industrie, 11 (1924), 557-566. 

2 Information from Forest Products Laboratory. 

3 P. Klason, Wochbl. Papierfabr., 41 (1910), 464-468, 541-546. 



396 



CHEMISTRY OF CELLULOSE AND WOOD 



which 2.10 per cent was free and 1.08 per cent combined. The 
spent liquor contained : 



CaO in 

solution, 

grams 


CaO as 
CaS0 4 , 
grams 


S0 2 as 

SO,, 

grams 


S0 2 as 

S, 
grams 


S0 2 

free, 
grams 


S0 2 in 
lignin, 
grams 


18.8 


23.7 


27.1 


13.5 


6.75 


90.4 



The presence of free stable ligninsulphonic acid or sulphuric 
acid causes resinification of the lignin. Sulphuric acid is formed 
by heating sulphurous acid, the reaction taking place rapidly- 
above 150°: 3S0 2 = 2S0 3 + S. Removal of the sulphuric acid 
by the lime promotes the above reaction, so that neither too 
much lime nor too much free S0 2 should be present. For the 
best cooking conditions, the free sulphur dioxide should be 70 
per cent of the total. Free sulphur is a catalyst for the formation 
of sulphuric acid from sulphur dioxide, though selenium is 500 
times more powerful. 1 The activity of selenium is due to the 
simultaneous presence of iron oxide carried over during the 
burning of seleniferous pyrites. 2 

The reactions involving the removal of lignin are intimately 
associated with the hydrogen-ion concentration of the sulphite 
liquor. 3 The P H of the average sulphite liquor is about 1.5. 4 
On heating, this value drops rapidly to 1.3. In order to obtain 
a pulp of maximum strength, p H 1.2 should not be exceeded. 

Oman 5 states that the greater the hydrogen-ion concentration, 
the darker is the color of the sulphite liquor. Under conditions 
obtaining in the usual sulphite cook, the color depends on the 
concentration of sulphurous acid in excess of that required to 
form calcium bisulphite, and is otherwise independent of the 
amount of lime present. According to Hagglund, 6 the formation 
of dark liquors does not necessarily mean a dark pulp, nor can it 



1 P. Klason, I.e. ; P. Klason and H. Mellquist, Papier-Fabr., 11 (1913), 
145-148. 

2 J. C. Torgersen and C. Bay, Papier-Fabr., 12 (1914), 483-484. 

3 P. Klason, Svensk Kern. Tid., 33 (1921), 141. 

4 R. N. Miller and W. H. Swanson, Unpublished data. 

5 E. Oman, Chem. Ztg., 39 (1915), 820. 

6 E. Hagglund and C. B. Bjorkmann, Svensk Kern. Tid., 36 (1924), 140. 



PULP PROCESSES AND WOOD PULPS 397 

be assumed that the two phenomena have the same origin. The 
dark color of the liquor shows that all the calcium bisulphite 
has been consumed, while the dark color of the pulp indicates, 
in addition, that the hydrogen-ion concentration has reached a 
certain value. Free ligninsulphonic acid and sulphuric acid are 
then present in considerable amounts. The formation of sul- 
phuric acid is not understood, though copper, iron, and lead are 
potent catalysts. 

In a normal cook, the formation of sulphuric acid is relatively 
insignificant. Harpf 1 found only a slight amount of calcium 
sulphate in the finished cook. Miller and Swanson, 2 using a 
raw sulphite liquor containing 0.11 per cent sulphuric acid, 
obtained 0.28 per cent of sulphuric acid in the spent liquor from 
a cook lasting 15.75 hours. The formation of sulphuric acid 
appears to be independent of temperature, but related to the 
concentration of free sulphur dioxide. 

In view of the action of strong acids on lignin, a proposed 
variation in the sulphite process is of interest. This consists 
in boiling the wood with 1 per cent sulphuric acid at 115° for 
2 to 4 hours before heating with sulphite liquor. 3 Waentig 4 
found that if fine shavings were subjected to the acid treatment, 
then chlorinated to remove the lignin, there was obtained 41 to 
42 per cent of cellulose resembling hydrocellulose in its brittle- 
ness. Chips, on the other hand, gave 48 per cent of long, strong 
fibers. The difference is attributed to transverse cutting of the 
fibers in the shavings, whereby the interior of the cells was 
attacked. Whether boiled with acid or not, the cellulose from 
both shavings and chips contained uniformly 84 to 85 per cent of 
a-cellulose. Sulphite pulp after heating with acid showed only 
74.3 per cent of a-cellulose. The cellulose from the acid-treated 
wood, when converted into viscose, showed a decidedly lower 
viscosity. The heterogeneous character of massive wood appears 
to protect it from brittleness on treatment with acid, although 
the resulting cellulose manifests a higher chemical activity. 

1 A. Harpf, "Beitrage zur Kenntnis der chem. Vorgange beim Sulfitver- 
fahren," Diss. Bern (1892), p. 26. 

2 R. N. Miller and W. H. Swanson, Paper Trade J., 76, 10 (1923), 53. 
3 E. Bronnert, U. S. P. 1392047 (1921). 

4 P. Waentig, Zellstoff u. Papier, 2 (1922), 152-155. 



398 CHEMISTRY OF CELLULOSE AND WOOD 

By-products. — The volatile oil obtained in cooking coniferous 
wood by the sulphite process contains some Z-borneol, 1 but 
consists largely of cymene, as was first shown by Klason. 2 The 
yield varies from 0.36 to 1.0 gallon of oil per ton of pulp. 3 
According to Kertesz, 4 the oil contains 80 per cent of cymene, 
the remainder being a mixture of sesquiterpene and diterpene. 
These hydrocarbons were believed to be formed from terpenes 
and resins by the reducing action of the sulphite liquor, accom- 
panied by polymerization, isomerization, and dehydration. 
Cross and Engelstad 5 shared the general belief that cymene was 
formed by the direct action of sulphurous acid. A tube contain- 
ing aqueous sulphur dioxide and turpentine, heated at 110° for 
26 hours, showed the presence of sulphur. It was assumed 
that there was a slight conversion to cymene. The formation 
of cymene appears, however, to be due to oxidation by sulphur 
set free during the cooking process. Pinene, heated with 50 
per cent of sulphur at 200° for 23 hours, gave about 50 per cent 
of cymene. 6 The light oil obtained by the acid hydrolysis of 
wood under pressure contains only 4 to 5 per cent of cymene. 7 

In laboratory cooks, the yield of acetic acid varied from 2.6 
to 4.2 per cent of the weight of the wood, the average correspond- 
ing to one acetyl group per molecule of a-lignin. 8 The formation 
of formic acid was insignificant, being only 0.04 to 0.09 per cent. 
Honig 9 obtained 2.151 to 9.078 grams of volatile acids per liter 
of sulphite liquor, this being equivalent to 0.72 to 3.03 per cent 
of the weight of the wood. In three liquors examined, the 
ratio of acetic to formic acid varied from 5.6:1 to 13.6:1. The 
occurrence of succinic and protocatechuic acids has also been 
reported. 10 

1 P. Klason and B. N. Segerfelt, C. A., 8 (1914), 1345; H. Bergstrom, 
Papier-Fabr., 13 (1915), 229; A. S. Wheeler and C. R. Harris, J. Am. 
Chem. Soc, 47 (1925), 2836. 

2 P. Klason, Ber., 33 (1900), 2343. 

3 A. W. Schorger, J. Ind. Eng. Chem., 10 (1918), 258-260. 

4 Z. Kertesz, Chem. Ztg., 40 (1916), 945-948. 

6 C. F. Cross and A. Engelstad, J. Soc. Chem. Ind., 43 (1924), 255T. 
6 S. Komatsu et al, C. A., 17 (1923), 2577. 

7 E. Heuser et al, Z. angew. Chem., 36 (1923), 37-38. 

8 E. Hagglund and C. B. Bjorkmann, Svensk Kern. Tid., 36 (1924), 142. 

9 M. Honig, Chem. Ztg., 36 (1912), 889-890. 
10 W. Buddens, Papier-Zeit, 16 (1891), 1813. 



PULP PROCESSES AND WOOD PULPS 



399 



The high acetyl content of hardwoods may be the reason why, 
in cooking these woods by the sulphite process, it is necessary 
to use a liquor high in bases. This permits neutralization of the 
acetic acid which is so readily formed. 1 

The relief gases contain acetaldehyde, acetone, methyl alcohol, 
acetic and formic acids, and furfural. 2 One thousand liters of 
sulphite liquor, equivalent to 100 kilograms of pulp, will give: 

Kilograms 

Acetaldehyde . 06 

Acetone 0.20 

Methyl alcohol 0.53 

Ethyl alcohol (by fermentation) 6 . 32 

According to Bergstrom, 3 8 to 10 kilograms of methyl alcohol 
are produced per metric ton of pulp. About 3 kilograms escape 
in the relief gas. The yields from pine and spruce differ little. 

Spent sulphite liquor contains considerable amounts of reduc- 
ing sugars formed by hydrolysis of hemicelluloses and a portion 
of the cellulose. In some cases, appreciable quantities of unhydro- 
lyzed carbohydrates may exist in the solution. 4 Krause 5 found 
the following amounts of sugar: 





Mitscherlich, 
per cent 


Ritter-Kellner, 
per cent 


Total sugars 


1.48 
0.47 
0.48 
0.28 
0.01 


1.47 


Pentoses 


0.41 


Mannose 


0.48 


Fructose 


0.25 


Galactose 


0.01 


Glucose 


Trace 







According to Hagglund, 6 the liquor contains 2.3 per cent of 
sugar, consisting of: 

Per Cent 

Xylose 30 . 

Mannose 41.7 

Galactose 2.6 

Fructose 3.0 

Undetermined 22 . 7 

1 C. G. Schwalbe and E. Becker, Z. angew. Chem., 32 (1919), 231. 

2 Z. Kertesz, Chem. Ztg., 40 (1916), 945. 

3 H. Bergstrom, Papier-Fabr., 8 (1910), 506; 10 (1912), 677. 

4 E. Oman, Tecknisk Tids., 46 (1916), 6-8. 
6 H. Krause, Chem. Ind., 29 (1906), 217. 

6 E. Hagglund, Biochem. Z., 70 (1915), 416. 



400 



CHEMISTRY OF CELLULOSE AND WOOD 



It is possible to obtain up to 20 per cent of sugars from the wood 
without injuring the pulp. 1 

Klason 2 gives the sugar content of sulphite liquor as 3.3 to 3.6 
per cent. The composition of the sugars formed is compared 
with that of the wood in the following table: 





Wood, 
per cent 


Sugars formed based 
on wood 


Per cent 


Cellulose 


50.0 

16.0 

30.0 

0.7 

3.3 


Glucose 


7.9 


Hemicelluloses 


Mannose 


2.5 


Lignin 


Galactose 

Arabinose 


1 3 


Protein 


4 3 


Resin and fat 


Xylose 












26 
24 
22 
20 

18 

IG 

10 
8 
fb 

4 
2 

°0 2 3 4 5 G ^1 8 9 10 1112 13 14 

Hours 
Fig. 8. — Comparison of rate of removal of lignin and formation of sugar in a 
sulphite cook. {After Sherrard and Suhm.) 

On fermentation, the maximum yield of alcohol is 1.4 per cent 
of the volume of the liquor — usually only 1 per cent. 

1 E. Hagglund, Svensk Kern. Tid., 36 (1924), 292. 

2 P. Klason, Papier-Fabr., 15 (1917), 641; 16 (1918), 1; Svensk Papers 
Tid., 20 (1917), 176. 













































































>i 


7 
























\ 


ff 




















































i 


























J 










o 


ju_ 
























V x 


p 














J 


V 






$ 
















•r 






V 






















V 




<$ 
























y 


$ 
























y v 


f> 

















PULP PROCESSES AND WOOD PULPS 401 

Canadian sulphite liquors contain 0.5 to 1.75 per cent of 
fermentable sugars. 1 Owing to the higher concentration of 
sulphur dioxide in the cooking liquors in winter, the spent liquors 
contain 1.94 to 2.80 per cent of sugar in winter and 1.65 to 2.20 
per cent in summer. 2 

The rate of formation of sugars in the Mitscherlich process 
has been studied by Sherrard and Suhm. 2 The total concentra- 
tion of sulphur dioxide was 3.8 per cent, of which 2.39 per cent 
was free and 1.41 per cent combined. The first sugar determina- 
tion, made at the end of the fourth hour, showed that little 
sugar was formed below 100°. From this point, the rate of sugar 
formation increased with the steam pressure to about the seventh 
hour, and was greatest between the seventh and ninth hours 
(Fig. 8). At the end of the eleventh hour, the point corresponding 
with the best quality of pulp, the yield of sugar was 1 1 per cent. 
Increasing the total sulphur dioxide to 5.60 per cent, of which 
4.46 per cent was free, gave 16 per cent of sugar, but the increase 
was offset by a corresponding decrease in the yield of pulp. The 
rate of removal of lignin and the formation of sugar were equal 
up to the eighth hour, at which point the removal of the lignin 
became much more rapid. When 90 per cent of the lignin had 
been removed, the yield of sugar was 12 per cent. 

In the quick-cook process, the sugar is formed more rapidly 
and in greater yield than in the Mitscherlich process. 3 Forma- 
tion of sugar is rapid at the beginning of the third hour and is 
completed at the end of the seventh. The usual variations in 
the quantity of free sulphur dioxide in the digester have no 
appreciable effect on the quantity of sugars produced, nor do 
variations between 130 and 150° in the final temperature. 
Spruce gave about 25 per cent more sugar than jack pine. 

Before it can be fermented, the liquor must be neutralized. 
In addition to sulphur dioxide, which is very toxic to yeast, the 
solution contains 0.15 to 0.73 per cent of acetic acid, and 0.008 

1 B. Johnsen and R. W. Hovey, " Utilization of Waste Sulphite Liquor," 
Dept. Interior, Canada, Forestry Branch, Bull., 66 (1919), 105. 

2 E. C. Sherrard and C. F. Suhm, /. Ind. Eng. Chem., 14 (1922), 931; 
cf. E. Hagglund, "Die Sulfitablauge . . .," p. 18. 

3 E. C. Sherrard and C. F. Suhm, J. Ind. Eng. Chem., 17 (1925), 194; 
for the cooking conditions see R. N. Miller and W. H. Swanson, Paper 
Trade J., 78, 15 (1924), 178. 



402 



CHEMISTRY OF CELLULOSE AND WOOD 



Formation of Sugars from Spruce by the Quick-cook 
Sulphite Process 



Cook 


Time, hours 


Maximum 
temperature 


Sugar on 
wood, 
1 per cent 


Yield of 

pulp, 
per cent 


676 
677 
678 
680 

682 
683 


6.5 
6.0 
9.0 
8.5 
6.5 
6.5 


140° 

130 

131 

130 

140 

150 


17.97 
18.40 
16.82 
21.60 
17.80 
18.26 


46.1 

55.0 

47.35 

48.8 

46.0 

45.55 



to 0.016 per cent of formic acid. 1 Neutralization is partially 
accomplished with lime and completed by agitation with calcium 
carbonate. This requires 4 to 5 hours. Complete neutraliza- 
tion with lime produces partial destruction of the sugars and 
gives an undesirable alkaline reaction. After settling, the 
clear liquor is fermented at a temperature of 29 to 30°, for 70 
to 80 hours, by the addition of beer yeast and nutrients. The 
yield of alcohol averages 1 per cent by volume. 

It is preferable to have the liquor slightly acid. If alkaline, 
the hydrogen-ion concentration being between 10~ 8 and 10~ 9 , 
the sugar is split into glycerine and acetaldehyde, thus 
reducing the yield of alcohol: 

C 6 H 12 6 = C 3 H 8 3 + C0 2 + CH3.CHO. 
The acetaldehyde then undergoes the Cannizzaro reaction: 

2CH3.CHO + H 2 = C2H5OH + CH3.COOH. 
The composition of the crude alcohol is given below: 



I, 

Per cent 
by weight 



II, 

Per cent 
by weight 



III, 

Per cent 

by weight 



Strength of alcohol . 

Methyl alcohol 

Acetaldehyde 

Fusel oil 



94.2 
3.2 
0.35 
0.24 



94.0 
0.8 
1.0 
0.7 



94.5 
0.35 
0.00 
0.10 



1 E. Hagglund, "Die Sulfitablauge und ihre Verarbeitung auf Alcohol," 
Braunschweig (1921), 82 pp. 



PULP PROCESSES AND WOOD PULPS 403 

Acetone is present in very small amount. Acetaldehyde is 
usually present though the amount is variable. Two sam- 
ples of the crude alcohol contained 2.61 and 5.48 per cent of 
acetaldehyde, and 1.28 and 0.54 per cent of paraldehyde. 1 On 
fractionation, the portion of b.p. 21 to 22° contains 75 per cent of 
acetaldehyde; that between 22 and 50°, 30 per cent of acetalde- 
hyde besides esters; while the fraction between 50 and 70° contains 
mainly acetone, 20 per cent of methyl and ethyl alcohols, and 
small amounts of aldehydes. A fraction, b.p. 101 to 120°, was 
found to consist of almost pure acetal. 2 The "tails" consist 
mainly of fusel oil. There have been identified propyl, isobutyl, 
butyl, and amyl alcohols, and borneol. 

Soda Process. — It requires at least 20 per cent of sodium 
hydroxide to obtain a good pulp. In practice an excess of alkali 
is used, so that the amount varies from 25 to 35 per cent of the 
dry weight of the wood. A portion of the black liquor, on 
account of the free alkali present, is used for the next cook. For 
every cord of wood in the form of chips, there are used about 900 
gallons of sodium hydroxide of a concentration of 6 to 10 per 
cent. From 1 to 2 hours are required to bring the digester to 
the cooking temperature, where it is held 4 to 8 hours. Heating 
is performed by the introduction of steam. The steam pressure 
varies from 100 to 130 pounds, corresponding to temperatures 
of 170 to 180°. Circulation within the digester is obtained by 
releasing the pressure at intervals. 

The "black" or spent liquor is evaporated to a sp. gr. of 1.4 
to 1.6, incinerated, the ash causticized with lime and reused. 
Owing to mechanical losses, it is necessary to add about 20 per 
cent of fresh sodium hydroxide. 

The yield of pulp is about 40 per cent. Conifers require more 
alkali per cord, longer cooking, and give less pulp per cord than 
hardwoods. 3 Hardwoods are cooked in about 4 hours, while 
spruce requires 6 to 7 hours. The following data are the result 
of mill experience ; 

1 E. Heuser et al, Papier-Fabr., 20 (1922), 1-5. 

2 A. W. Owe, Papier-Fabr., 20 (1922), 1564. 

3 J. A. de Cew, /. Soc. Chem. Ind., 26 (1907), 561-563. 



404 CHEMISTRY OF CELLULOSE AND WOOD 

Alkali Required and Yield of Pulp 



Species 



NaOH as 
Na 2 C0 3 

pounds 



Wood 



Spe- 
cific 
grav- 
ity 



Weight 

per 

cord, 

pounds 



Pulp 



Oven- 
dry, 
per 
cent 



Air- 
dry, 
pounds 



Black spruce (Picea nigra) 

Hemlock (Tsuga canadensis) . . . 
Poplar (Populus grandidentata) 
Basswood (Tilia americana) . . . 

White birch (Betula alba) 

Soft maple (Acer rubrum) 

Yellow birch (Betula lutea) 



900 


0.41 


2,250 


40 


950 


0.42 


2,300 


38 


800 


0.43 


2,350 


44 


800 


0.425 


2,325 


44 


800 


0.58 


3,190 


42 


850 


0.64 


3,520 


40 


850 


0.66 


3,630 


40 



1,000 
970 
1,150 
1,135 
1,490 
1,560 
1,610 



Aspen may be successfully cooked with 20 to 25 per cent of 
sodium hydroxide. 1 Yields as high as 56 to 58 per cent can be 
obtained in a cooking period of 3 to 4 hours. 

Digestion of wood with caustic soda in the presence of oxygen 
lowered the yield of pulp from 41.5 to 38.2 per cent, while the 
bleach consumption increased from 9.0 to 11.3 per cent of 
chlorine. 2 Cooking is facilitated by applying a vacuum to the 
wood in the digester, allowing the cooking liquor to enter, and 
applying a pressure of 100 to 110 pounds. In this way the air 
is removed and a very uniform impregnation of the chips is 
obtained. Rinman 3 claims to improve the yield of cellulose by 
having a small amount of a catalytic reducing metal, such as 
mercury, present in the cooking liquor. 

During the cooking process, the carbohydrates passing into 
solution are converted into saccharinic acids, while the lignin is 
converted into soluble sodium salts. According to Hagglund, 4 
Norway spruce gives 42.8 per cent of cellulose, while the waste 
liquor contains 21.6 per cent lignin, 3.2 per cent acetic acid, 1.7 
per cent formic acid, 1.7 per cent methyl alcohol, and 18.2 per 

1 H. E. Surface, ''Effects of Varying Certain Cooking Conditions in 
Producing Soda Pulp from Aspen," U. S. Dept. Agr. Bull., 80 (1914). 

2 H. Wallin, C. A., 14 (1920), 3790. 

3 E. L. Rinman, U. S. P. 1319360 (1919); B. N. Segerfelt, Pulp Paper 
Mag. Can., 19 (1921), 452-453. 

4 E. Hagglund, Cellulosechemie, 5 (1924), 81-87. 



PULP PROCESSES AND WOOD PULPS 405 

cent of oxyacids ; of the latter, 55 per cent are present as oxyacids 
and 45 per cent as lactones. Lactic acid is present to the extent 
of 0.9 per cent of the weight of the wood. 
Pine wood cooked with caustic soda gives: 1 

Per Cent 

Pulp 37.0 

Turpentine 0.4 

Methyl alcohol 0.5 

Fats and resins . 75 

Humus and humic acids 31 . 00 

Soluble humic acid, oxyacids and lactones 20.00 

Acetic acid 3 . 00 

Formic acid 3 . 00 

The humus substances may be precipitated by saturating the 
black liquor with carbon dioxide at 75°, or at proper concentrations, 
with sodium chloride and other salts. The humic acids are not 
affected by this treatment, but may be precipitated with acids. 
The oxyacids contain lactic acid and saccharinic acids. 

Bergstrom 2 cooked wood by the soda process at a pressure of 
140 pounds, and obtained the following yields of methyl alcohol 
based on the dry wood ■ 

Methyl Alcohol, 
Per Cent 

Spruce (Picea excelsa) . 67 

Pine (Pinus silvestris) . 67 

Longleaf pine . 68 

Shortleaf pine . 66 

Poplar 0.67 

White birch 0.81 

Eucalyptus . . 83 

The distribution of methoxyl in the products from cooking 
jack pine (Pinus divaricata) with sodium hydroxide has been 
investigated by Aiyar. 3 Each charge consisted of 45.4 kilograms 
of wood and 20 per cent of sodium hydroxide having a concentra- 
tion of 96 grams per liter. The wood contained 4.84 per cent 
of methoxyl, equivalent to 2.2 kilograms. After the first 2 
hours the removal of methoxyl was very slow. Data on the 
amounts of lignin and methoxyl in the pulps indicate that all 

1 E. L. Rinman, Papier-Fabr., 10 (1912), 39, 101. 

2 H. Bergstrom, Papier-Fabr., 11 (1913), 427-428. 

3 S. S. Aiyar, J. Ind. Eng. Chem., 15 (1923), 714-716. 



406 



CHEMISTRY OF CELLULOSE AND WOOD 



the methoxyl is associated with the lignin, and that the methoxyl 
is combined with the lignin in only one way. 

Distribution of Methoxyl in Cooking Jack Pine 



Duration 

of cook 

from 

maximum 

pressure, 

hours 


Yield 
of 

pulp, 
per 
cent 


CH 3 in pulp 


CH3O in black 
liquor 


CH3O in black 
liquor 


Kilo- 
grams 


Per 

cent 


Kilo- 
grams 


Per 

cent 


Fixed, 
kilo- 
grams 


Volatile, 
kilo- 
grams 


-0.5 


+0.5 

1 

2 

3 

4 

6 


80.7 
66.5 
58.4 
51.6 
44.6 
46.0 
37.1 
45.7 


1.714 

1.196 

0.914 

0.64 

0.33 

0.32 

0.33 

0.24 


4.67 
3.96 
3.45 
2.74 
1.60 
1.52 
1.95 
1.19 


0.59 

1.127 

1.40 

1.83 

1.89 

1.92 

2.70 

2.34 


0.43 

0.46 

0.567 

0.985 

0.975 

1.180 

0.980 

0.927 


0.22 
0.84 
0.64 
1.27 
1.50 
1.64 
1.64 
2.26 


0.45 
0.43 
0.78 
0.76 
0.63 
0.47 
0.94(?) 



A series of soda cooks 1 made on aspen, loblolly pine, and jack 
pine permits following the changes in the composition of the 
wood as the cook progresses. The woods were cooked with 
20 per cent of their weight of sodium hydroxide at a maximum 
temperature of 170° over periods of 0.5 to 7 hours. The chips 
were impregnated with the liquor at a pressure of 100 to 110 
pounds per square inch, at 20 to 30°, before cooking. The data 
show that approximately 90 per cent of the cooking takes place 
during the first 2 hours. As judged by the chlorine method for 
determining cellulose, the cellulose is initially removed faster 
than the lignin. This is due mainly to loss of pentosans. In 
the case of aspen, the pulp retains uniformly about 8 per cent of 
very resistant pentosans. Only in the middle of the cook is the 
lignin removed faster than the cellulose. Most of the acetic 
and formic acid is produced during the first half hour. The total 
organic solids in the black liquor reach a maximum at the end 
of the first 2 hours. 

Many attempts have been made to obtain by-products from 
the black liquor. Destructive distillation of the black liquor 

1 S. D. Wells et al, Paper Trade J., 76, 24 (1923), 49-55; M. W. Bray 
and T. M. Andrews, Ibid., 76, 19 (1923), 49-51. 



PULP PROCESSES AND WOOD PULPS 



407 



from a cord of hardwoods gave 6.9 gallons of a mixture of methyl 
alcohol and acetone. 1 Binman 2 mixed the spent liquor with 
lime and distilled it at a temperature of 400° in the presence of 

Data on Soda Cooks 







Based on dry wood 


Cellu- 
lose 
in 
pulp, 


Species 


Time of 
cook, 
hours 


Yield of 
pulp, 


Lig- 
nin, 


Cellu- 
lose, 


a-Cellu- 
lose, 






per 


per 


per 


per 


per 






cent 


cent 


cent 


cent 


cent 







100.0 


23.4 


62.1 


43.2 


62.1 




0.5 


51. 5(?) 


18.9 


57.0 


37.3 


71.0 




1.5 


60.5 


4.8 


55.0 


52.2 


91.0 


Aspen 


•i 


57.5 
55.0 


3.2 
1.4 


53.6 
53.3 


40.8 
41.4 


83.3 




97.0 




4 


52.5 


0.9 


51.0 


40.0 


97.1 




5 


52.0 


0.8 


50.4 


37.0 


96.9 




>7 


49.4 


0.7 


48.0 


39.7 


97.4 







100.0 


29.7 


60.5 


24. 2(?) 


60.5 




0.5 


81.4 


25.2 


52.4 




64.4 




1.5 


70.3 


16.6 


51.4 




73.1 




2 


61.2 


12.2 


47.2 




77.1 




< 3 


49.4 


5.3 


43.2 


25.8 


87.5 


Loblolly pine 


4 


49.0 


4.7 


43.2 


28.6 






88.0 




5 


47.8 


4.4 


44.4 


34.3 


92.8 




6 


41.5 


1.9 


38.2 


25.6 


93.1 




} 7 


39.9 


1.6 


37.4 


25.6 


93.6 




) 


100.0 


32.8 


59.8 


42.6 


59.8 




0.5 * 


80.7 


25.6 


50.4 


16. 5(?) 


62.5 




1.5J 


58.4 


14.0 


43.0 


32.6 


73.6 


Jack pine 


< 2 


51.6 


9.0 


42.0 


35.1 


81.5 




3 


43.8 


3.2 


41.0 


28.5 


93.7 




4 


39.9 


2.6 


37.8 


27.4 


94.8 




5 


37.0 


1.0 


35.6 


29.7 


96.4 



* Old liquor. 

steam. There were obtained per ton of pulp, 400 to 500 kilo- 
grams of fuel gas, 100 kilograms of higher alcohols, ketones, 
hydrocarbons, cresols, etc., and 100 kilograms of a spirit con- 
sisting of 50 per cent acetone, 25 per cent methylethyl ketone, 

1 A. H. White and J. D. Rue, Met. Chem. Eng., 16 (1917), 182-186. 

2 E. L. Rinman, Papier-Fabr., 10 (1912), 39-41, 101-104. 



408 CHEMISTRY OF CELLULOSE AND WOOD 

and 25 per cent of ethyl alcohol. Little methyl alcohol was 
produced. 

Sulphate Process. — The sulphate process differs from the 
soda process through the presence of a certain amount of sodium 
sulphide, the sulphur of which is lost for the most part during 
the cook and the subsequent treatment of the spent liquor. The 
name is derived from the custom of restoring the sulphur by the 
addition of sodium sulphate to the black liquor previous to 
incineration, sodium sulphide being formed by reduction. A 
modification is the soda-sulphur process, wherein sulphur is 
added directly to the caustic soda liquor. During cooking some 

6NaOH + 4S = 2Na 2 S + Na 2 S 2 3 + 3H 2 

sodium thiosulphate is formed in addition to sodium sulphide. 
The pulps obtained by the sulphate process and its modification 
are essentially identical. 

The sodium sulphide has a very beneficial effect not only on 
the strength of the pulp but on the yield as well, which is 10 
per cent greater than by the soda process. The function of the 
sulphur is not completely understood. According to Klason 
and Segerfelt, 1 the sodium sulphide does not become caustic 
until the sulphur has entered into combination with the lignin, 
so that the cellulose is not subjected to as great actual causticity 
as in the soda process. Arrhenius 2 has determined that the 
digestion of wood with sodium sulphide, sodium hydroxide, or 
their mixture, can be expressed as a unimolecular reaction. The 
incrustants are dissolved 28 times faster than the cellulose at 
148°, but only 8.7 times faster at 170°. At 270°, the cellulose and 
incrustants dissolve at the same rate. With sodium sulphide 
and sulphate liquors, the lignin is attacked somewhat more 
rapidly than with caustic soda, while the velocity of attack on 
the cellulose itself is greatly reduced. 

Wells 3 cooked jack pine with 20 per cent of sodium hydroxide 
and 2 per cent of sulphur at a temperature of 170°. The sulphur 
accelerated the removal of incrustants and reduced the loss of 
cellulose during the first half hour, 50 per cent. The papers 

1 P. Klason and B. Segerfelt, Papier-Fabr., 9 (1911), 1093-1099. 

2 S. Arrhenius, Zellstoff u. Papier, 4 (1924), 182-184. 

3 S. X>. Wells, Pulp Paper Mag. Can., 21 (1923), 623-626. 



PULP PROCESSES AND WOOD PULPS 



409 



made from the sulphate pulp showed marked increase in bursting 
and tensile strength, and in tearing and folding resistance over 
soda pulp. From a chemical standpoint there is little difference 
between soda and sulphate pulps. 

Data on Sulphur-soda Cooks 



Cook 
No. 



Time to 
maxi- 
mum 

temper- 
ature, 

minutes 



Total 

time 

of cook, 

minutes 



Based on dry wood 



Yield 
of pulp, 
per cent 



Lignin, 
per cent 



Cellu- 



per cent 



a-Cellu- 

lose, 
per cent 



Cellu- 
lose in 
pulp, 
per cent 



Wood 
865 
866 
867 
870 
868 







100.0 


32.65 


58.9 


40 


40 


82.75 


24.8 


54.6 


70 


70 


63.0 


13.5 


47.4 


60 


105 


48.6 


4.0 


45.8 


60 


150 


40.55 


2.1 


38.6 


85 


173 


43.2 


2.5 


41.0 


70 


246 


43.03 


2.0 


41.6 



42.2 
36.3 
35.7 
37.8 
30.0 
32.0 
34.4 



58.9 
65.9 
75.2 
94.1 
95.1 
94.9 
96.8 



As in the soda process, the carbohydrates are largely converted 
into oxyacids. Based on the organic matter, the spent liquor 
has the following composition: 1 

Per Cent 

Lignin 54.29 

Resin and fatty acids 2 . 47 

Formic acid 3 . 69 

Acetic acid 5.16 

Oxyacid lactones 30 . 34 

Meta- and para-saccharinic acids were identified, though the 
principal constituent is an isomer of isosaccharinic acid. 

The chemicals used in the sulphate process represent about 20 
per cent, calculated as sodium hydroxide, of the weight of the 
wood. The liquor used in the digesters contains 90 to 100 grams 
of sodium hydroxide, 25 to 30 grams of sodium sulphide, and 10 
grams of sodium carbonate per liter. The wood is cooked at a 
temperature of 150 to 170° for a period of 3 to 6 hours. The 
yield of pulp varies from 45 to 50 per cent. 



P. Klason and B. Segerfelt, I.e. 



410 CHEMISTRY OF CELLULOSE AND WOOD 

A regenerated soda ash had the following composition: 

Pee Cent 

Sodium carbonate 61 . 73 

Sodium sulphide 21 . 50 

Sodium sulphite and thiosulphate 7 . 33 

Sodium hydroxide 3 . 50 

Sodium silicate 1 . 22 

Sodium sulphate 2 . 78 

Ferric oxide . 04 

Insoluble and loss 1 . 90 

Of the sulphur originally present as sodium sulphide, 51.8 
per cent was combined with the lignin, 15 per cent was obtained 
as volatile sulphur compounds, principally mercaptan and 
dimethyl sulphide, and 16.4 per cent was unaccounted for. The 
volatile sulphur compounds, particularly the mercaptan, have 
an exceedingly unpleasant odor. Under the same conditions, 
pine wood gave twice as much mercaptan as spruce. The 
amount of mercaptan formed can be reduced, and the dimethyl 
sulphide increased by increasing the amount of alkali, whereby 
also an easy-bleaching pulp is obtained. In this case, about 100 
grams of mercaptan are formed from a ton of dry wood. If the 
amount of alkali is reduced in order to obtain a kraft pulp, the 
amount of mercaptan formed may be increased tenfold. During 
the cooking operations, 20 per cent of the methoxyl groups are 
split off the lignin as methyl alcohol, mercaptan, and dimethyl 
sulphide. 

The formation of methyl alcohol and its derivatives takes 
place as follows: 1 

R.OCH 3 + NaOH = R.ONa + CH 3 OH. 
R.OCH3 + NaSH = R.ONa + CH 3 SH. 
CH 3 SH + NaOH = CH 3 SNa + H 2 0. 
R.OCH3 + CH 3 SNa = R.ONa + (CH 3 ) 2 S. 

The non-condensable gases from a digester producing 1375 
kilograms of pulp, contained about 20 grams of mercaptan. The 
condensate contained 2 liters of oil, of which 200 grams was 
dimethyl sulphide, the remainder turpentine. The condensed 
water contained 5 kilograms of methyl alcohol smelling strongly 



P. Klason, Wochbl. Papierfabr., 39 (1908), 3859-3861, 3963-3966. 



PULP PROCESSES AND WOOD PULPS 



411 



of trimethylamine. The black liquor from 16.5 tons of pulp 
would produce, by destructive distillation, 4.2 kilograms of hydro- 
gen sulphide and 380 kilograms of mercaptan. During incinera- 
tion to recover the soda, 1.4 kilograms of hydrogen sulphide and 
4.5 kilograms of mercaptan escaped combustion and passed out in 
the chimney gases. The other sulphur losses in the flue gases 
were 2197 kilograms of sodium sulphite and 278 kilograms of 
free sulphur dioxide. 

According to Falk, 1 the relief condensate from the manufacture 
of a ton of cellulose from pine wood contains: 



In oil, 
kilograms 



In water, 
kilograms 



Mercaptan , 

Dimethyl sulphide 

Dimethyl disulphide 

Turpentine 

Residue from distillation 

Methyl alcohol 

Ammonia 



0.062 
0.927 
0.103 

8.487 
0.721 



0.06 
0.17 
0.05 
0.92 

5.00 
0.18 



Pine and spruce give about the same amount of crude methyl 
alcohol. 2 About 13 kilograms of methyl alcohol, containing 0.5 
per cent of acetone, are formed in the manufacture of 1000 
kilograms of pulp. 

None of the numerous methods proposed for disposing of 
the offensive relief gases is an unqualified success. Schwalbe 3 
passes the gases through comminuted wood, which acts as an 
absorbent or contact substance, the gases being destroyed by 
oxidation. After losing its efficiency, the wood can be cooked 
with alkali without producing the odor of mercaptan. The 
oxidation may be assisted by the use of chlorine or other oxidiz- 
ing agents. Klason 4 recommends passing the relief gases through 
efficient condensers, the uncondensed portion being burned in a 



1 H. Falk, Papier-Fabr., 7 (1909), 469-472. 

2 H. Bergstrom, Papier-Fabr., 10 (1912), 677-678. 

3 C. G. Schwalbe, Zellstoff u. Papier, 1 (1921), 69-72; 2 (1922), 175-178; 
G. P. 319594 (1917). 

4 P. Klason, Paper Trade J., 79, 2 (1924), 30-36. 



412 CHEMISTRY OF CELLULOSE AND WOOD 

furnace or absorbed in the concentrated black liquor before it is 
incinerated. Mercaptan gives with salts of the heavy metals 
the compound M // (CH 3 S) 2 ; however, only the mercaptides of 
the precious metals are odorless. 

It is difficult to remove the sulphur compounds from the 
turpentine obtained by the sulphate process so that it will 
remain odorless on prolonged standing. Agitation with 5 to 
10 per cent of 50 per cent sulphuric acid in the cold, until the 
odor disappears 1 , is only a temporary success. The best results 
are obtained with a solution of bleaching powder. 2 Other oxidiz- 
ing agents, such as oxides of nitrogen, 3 ultra-violet light in 
presence of air or oxygen, followed by ferric chloride, 4 have also 
been proposed. The logical though difficult step is to collect 
the turpentine before the temperature within the digester is 
sufficiently high to form organic sulphur compounds. 5 Spruce 
gives 1 to 1.5 kilograms and pine up to 10 kilograms of turpentine 
per ton of pulp. 6 When purified, the oil consists mainly of 
a-pinene and a small amount of /3-pinene. 

Sodium Sulphite Process. — In recent years, interest in the 
use of sodium sulphite as a pulping agent has been revived. 7 In 
the Keebra process, the wood is cooked with about 40 per cent 
of its weight of sodium sulphite. 8 The spent liquor is incinerated 
as in the sulphate process, during which principally sodium 
carbonate, sodium sulphide, and sodium sulphate are formed. 
Under certain conditions, a large part of the sulphur escapes 
in the flue gases as sulphur dioxide. The recovered ash is 
treated with sulphur dioxide, whereby sodium sulphite is 
reformed. 9 Sodium thiosulphate is also produced, owing to the 
presence of sodium sulphide. 

1 P. Klason and H. Person, Papier-Ztg., 32 (1907), 4; 33 (1908), 3779. 
2 B. R. Armour, E. P. 194286 (1923); C. A. Jobson, U. S. P. 1493454 
(1924); H. W. Fosse, Ber. pharm. Ges., 25 (1915), 303-313. 

3 A. Luttringer, Papier-Fabr. Fest u. Auslandsheft, 12 (1914), 64-67. 

4 H. Dedichen and O. M. Halse, U. S. P. 1253793 (1918). 

5 O. Rodowski, Paper Trade J., 78, 14 (1924), 75. 

6 O. M. Halse and H. Dedichen, Ber., 50 (1917), 623-630. 

7 W. Clifford, Paper Ind., 3 (1921), 1247; Anon., Paper Trade J., 77, 
10 (1923), 51. 

8 L. Bradley and E. P. McKeefe, Can. P. 219557 (1922). 
9 L. Bradley and E. P. McKeefe, Can. P. 236532 (1923). 



PULP PROCESSES AND WOOD PULPS 413 

The use of neutral sodium sulphite was protected by Cross 1 in 
1880, but held as of only theoretical interest. 2 In addition to 
sodium sulphite, 3 ammonium sulphite, 4 a mixture of sodium 
sulphite and sodium acetate, 5 and a mixture of sodium sulphite 
and sodium hydroxide 6 have been suggested. 

The Keebra process is unique in the use of a liquor that is 
nearly neutral, and this accounts for the unusual qualities of the 
pulps. Little information is available on the exact nature of 
the cooking process. The reactions involving the removal of 
the incrusting substances by means of sulphites and alkalis 
become antagonistic in a mixture of these chemicals. Replace- 
ment of 10 per cent of the sodium sulphite with sodium hydroxide 
greatly retards pulping. This appears to hold up to the point 
where 80 per cent of the mixture is represented by sodium 
hydroxide. The results are then similar to the soda process, 
though there is an improvement in the yield and quality of the 
pulp. 

The yield of pulp from coniferous woods is about the same as 
by the sulphate process, though about 25 per cent more chemicals 
are required. The pulp approaches sulphate kraft in strength 
and is much lighter in color, in which respect it resembles 
unbleached sulphite. Sodium sulphite gives with hardwoods 20 
per cent more pulp than the soda process. The pulp is excep- 
tionally strong, but lacks the softness, opacity , and felting proper- 
ties of soda pulp. 

Minor Pulping Processes, — The use of chlorine for removing 
lignin has also been revived. The process is essentially the Cross 
and Bevan laboratory method for determining cellulose. De 
Vains 7 uses an aqueous solution of chlorine, and Cataldi, gaseous 

1 C. F. Cross, E. P. 4984 (1880). 

2 C. F. Cross and E. J. Bevan, "Paper-Making" (1900), 66. 

3 W. Schacht, G. P. 122171 (1901) and 131118 (1902); Papier-Ztg., 26 
(1901), 3143; Papier-Fabr., 54 (1923), 2851. 

4 C. F. Sammet and J. L. Merrill, U. S. P. 1016178 (1912); L. Brech 
and E. Tyborowski, U. S. P. 1100519 (1914). 

5 M. Muller and O. Heigis, G. P. 284681 (1914); E. P. 156512 (1921). 

6 R. Blitz, F. P. 155014 (1883); L. Bradley, and E. P. McKeefe, 
Can. P., 246477 (1925). 

7 A. R. de Vains, E. P. 189561 (1921); 198975 (1922). 



414 CHEMISTRY OF CELLULOSE AND WOOD 

chlorine. 1 The raw material is boiled with alkali before chlorina- 
tion and the chlorinated derivatives are removed with alkali. 
According to Cerruti, 2 the preparation of 100 kilograms of dry- 
pulp from poplar by the Cataldi process requires a minimum of 
45 kilograms of chlorine and 9 kilograms of sodium hydroxide. 
Pomilio 3 gives a consumption of 28 kilograms of chlorine and 5 
kilograms of sodium hydroxide. The pulp from poplar contains 
74 per cent of a-cellulose and 21 per cent of /?- and y-cellulose. 

The use of chlorine gas on straw gives pulps having a high 
copper number, which reaches 9 in some cases. 4 Waentig 5 
considers chlorine gas no more difficult to use than chlorine water. 
Franz 6 has protected the use of a solution of chlorine in carbon 
tetrachloride. 

Chlorine has been used successfully in pulping the cereal straws, 
but its application to wood, in the form of chips, is hindered by 
the difficulty of securing penetration. Waentig and Gierisch 7 
boil wood shavings 0.2 to 0.3 millimeter thick with 0.1 per cent 
sodium hydroxide solution. The shavings are then treated in 
the cold with chlorine gas for several hours, washed to remove 
the hydrochloric acid formed, then treated with 1.0 per cent 
sodium hydroxide solution. Wood in the form of chips must be 
chlorinated under a pressure of 6 atmospheres. 

The most promising procedure consists in giving the wood a 
preliminary cook with sodium hydroxide or sodium sulphite, so 
that the wood can be mechanically separated into individual 
fibers before chlorinating. 

The use of phenols, 8 with or without the presence of a mineral 
acid, 9 is too expensive. The recovery of phenol is low, owing 

1 For a bibliography of the chlorine processes see C. J. West, Paper Trade 
J., 79, 14 (1924), 43-44. 

2 C. Cerruti, Giorn. chim. ind. applicata, 4 (1922), 64-65. 

3 U. Pomilio, Chimie & Industrie, 6 (1921), 267; 8 (1922), 41. 

4 W. Schacht, Papier-Fabr., 21 (1923), 521; 22 (1924), 121. 

5 P. Waentig, Papier-Fabr., 22 (1924), 456. 

6 A. Franz, G. P. 323936 (1919). 

7 P. Waentig and W. Gierisch, Text Forsch., 2 (1920), 69-79. 

8 F. A. Buhler, Chem. Ind., 26 (1903), 138; Papier-Ztg., 25 (1900), 3256; 
E. P. 6651 (1898). 

9 R. Hartmuth, G. P. 326705 and 328783 (1919); E. Legeler, Cellulose- 
chemie, 4 (1923), 61; C. Claessen, E. P. 160482 (1919). 



PULP PROCESSES AND WOOD PULPS 



415 



to its condensation with the lignin. The reaction is discussed 
in detail in the chapter on lignin. 

The use of aliphatic and aromatic amines, 1 and acid chlorides, 
such as sulphuryl chloride, thionyl chloride, acetyl chloride, 2 
etc., is of only general interest. It is stated that lignin can be 
removed from wood by heating it with concentrated formic or 
acetic acid, containing 0.3 to 0.7 per cent sulphuric acid, without 
injuring the cellulose. 3 

Wood cooked with calcium polysulphide 4 is not reduced to a 
pulp, but is softened sufficiently so that it can be separated into 
fibers mechanically. The reaction is accompanied by a disagree- 
able odor exceeding that of the sulphate process. 

Wood Pulps. — It is difficult to classify wood pulps on chemical 
data alone (Tables 1 to 4). Pulps showing nearly identical 
chemical composition may behave very differently from the 
standpoint of paper making. Krull 5 has used the terms u hard," 
" normal," and "soft" to define sulphite pulps. The classifica- 
tion is based on the lignin content, which was determined by 
treating the moist pulp with hydrogen chloride gas. A hard pulp 
is one suitable for the manufacture of pergamyn paper, and a soft 
pulp is particularly suitable for bleaching. A normal pulp has 
intermediate lignin values and is adapted to the manufacture of 
ordinary papers. 

Classification of Sulphite Pulps 



Mitscherlich 


Ritter-Kellner 


Lignin, per cent 


Designation 


Lignin, per cent 


Designation 


6.5-7.0 
5.5-6.0 
5.0-5.5 
1.0-2.0 


Hard 
Normal 
Soft 
Bleached 


7.5-9.5 
6.5-7.5 
4.5-6.5 


Hard 

Normal 

Soft 



1 W. Schlosser, G. P. 328729 (1919). 

2 R. Schwarzkopf, G. P. 328730 (1919). 

3 Akt. Ges. fur Zellstoff-u. Papier-Fabr., G. P., 309551 (1916); A. Foulon, 
ZeUstoffu. Papier, 5 (1925) 212-213. 

4 A. Tingle, Paper, 34 (1924), 653, 1000; V. Drewsen, U. S. P. 996225 
(1911). 

6 H. Krull, Papier-Fabr. Fest u. Auslandshefl, 19 (1921), 65-70. 



416 



CHEMISTRY OF CELLULOSE AND WOOD 



« 2 
g 03 

§1 

H $-. 

l-H CD 

CO Q, 



03 U 



s 5 



>> O 3 

W ° 5 



s si 

ft-Q "5 

1-1 CD 



U 



a -2 
a S 
° a 



NMOhOOO 

d d <-! d d d 



NMINO 

d d d d 





O 


o 


CO 












CO 


t^ 


o 






co 


co 


CM 


§ 












■* 


CO 


CD 












o 


00 


iC 


lO 




CN 


"tf 


iO 












US 


GO 


CO 












CO 


iO 


lO 


tH 




TH 


t^ 


CO 


lO 










h- 


CO 






OS 


co 




o 


CO 






co 


CO 


r^ 


co 




on 


Ol 


Ttl 


•^ 


1^ 


lO 


co 


co 






CO 


i« 


C3 


CO 


■-J 






CO 


CO 


'"I 


CN 




■~! 


id 


CM 


CO 


co 


o 



>0 TjH CO -* 



rtf CO CO K) lO >o 



(M ^ • • H N ^ O0 N • ■ 00 H O lO • ■* O O! O CO lO 
H ifl . ■ m N ■* O lO • • N N lO Tf • "# O 00 CO ■**< O 


COIN • •COCOCOCOCO • • CO CN CO CN • N N H CO m N 


HOOO)10NOMMCO(»N»OOHHNT|(00(HMN 

COiO(OHNOO^CC^O(NMiOOO^O(0«000)OOCO 



o o o o o 



oooooooooooo 



CO "0 CO CO 
CO Oi CO Ci 



CM CN <N <N 



HNNNNNNHNNNMNNMnncONHINO 



OOOOOOt-h^hOOOOOOOOOOOOOO 



T3 03 

533 



a a a a a 

d ^ ti ^ ^ u 
M M 00 M M oj 



-a -a t c 

a) rt cd 03 03 o3 

-a 32 -a -a .a ^ 



t3 ^ . ^-; 

03 CD . C 

e3 o3 -a S 



u v c 



T3 +J +i - 



03 03 0J 03 03 



03 3 £i 3 3 3 J 



a) a o j & a a 



ft -o u - 



3 a a a 



-3 -5 ^ a a 3 3 3 



oooowoowo 



CD 03 03 03 03 03 

J 3 J3 J J j 



H H R W W H 



« 03 03 



03 03 03 03 



0303030303030303^^0 



MM- . 

7 ^ -g ^ •« 

i ® " 



3 3 3 3 

OOOOOC300C3C3 

p ,* .■*" .^ .* * ^ ^ .^ ■" s * ? s * * s s ►• - 



PULP PROCESSES AND WOOD PULPS 



417 



w 53 



S 
o 
O 

o 

o 

5a 

o 

I 
o 

I 



<v 






















o 


00HNO5Q0NO)tDM 






^ "3 


MTfNKlCOMMHm 










<v 




O 




<p 




co 




j2 


NHHTflN^WHtO 


CD 

O 


MiONCOMHCO'JiCO 




CD 




O 


HOOtONOONOO'C'* 






k | 


CNNtONNWWitH 


t»aO0OO00O00GO0000 


0) 




• O 




CD 




O 


HKimtooooo^^H 






3 


(DiOifl^^OIMOON 




0>010)©0>0>0)0ffi 


o 




o 




a 




"S 


ONMONOOTfOHO 


to 


HiOHNCNHOO 


J 




-a 




ffl ® »H 


cr>C33C»cN(NOI>TtH«0 


0^0 H OOi HC OCOH 


b ° 








o *** 




o 




S 15 




"S gW 


lOiO^^OONNOCO 


ffl 3 h 9 


ONCNHOCN'tMH 


p » ^ 




| a * 




CO iH 




° *> »H 




2 o « 


OiffltONfflNWiOO 


~ .s £ 

co ~* 


d i-i d 1-! d h d o d 


CO cu 






s s as 




03 






CO 


CD CO JO 




*3 




O 


X2 


x> h xi 


T3 


~0 T3 T5 


O 


C 


a . a a 


£ 


c3 


03 -^ C3 03 




V 


s t) d i> o v " 




e 


O O t> £> r-5 eg w 




3 


3 3 p 3 g 3 3 




t-C 


a a a a J2 a a 




a 




CO ' 03 tB GO tB W CQtB 


III 






















o 
































a 


T3 


73 73 T3 T3 T3 






49 


Q3 4j Q3 dJ Q3 






•c 


_c 


j3 -a j3 ^ -a -o -d 


o 


o 


o o « u o 5 5 

c3 03 ct g3 03 ri r! 
4) OJ 4) Cv CU "S "S 


co 


o3 


CD 


0) 


Q 






X2 


J3 XJ X> -Q X3 03 =3 




a 


o a fl fl a j2 .2 




D • £ P U> £> £ pq pq 


Record 
num- 
ber 


©00tDMf3©Wt>"5 


t> 00 


a 


a 


Sv 


1- 

p- 


p. 


a 


p- 




P- 


i- 
P- 





418 



CHEMISTRY OF CELLULOSE AND WOOD 







o fl 
43 a> 
Q 2 


00 


N 
















■* 






CD O 












CD 




.fl 




i-H IN 












©' 




;£ 


< fl 


















* 






































o 


8-3 


oooaoN 












CN •* 




OS tJH ■* iO r-t IN 












iO <N 




X 


3 M 


o o o o o o 












o o 




|z| 


oj 


tN CN rt< iO O IO 












• iO 






xji i* N H H N 












• o 






W 


O O O O O O 












• o 




1 - on fl 

' D 5 T3 o! 

13 § 2 o > 
O ~ O * ffl 


• • 00 CO •* -co 












O • f~ 






• CN CO T* -CO 












<N • 05 






■ IO iti CO -00 
• OS OS OS -00 












iO • iO 
OJ • OS 


p 


















. fl <» 


M O) 

00 CO 






CO 








8 : 


Ph 


o 3 o 


CO 
















io' • 


00 


w 


o ■"■ 


00 00 • • < • 






00 






• 00 • 


H 


















«j 




















>> « 


H 00 *N N H^ 


OJ 




o 






00 Tt< ■* t> 


5 a 


CN i-H CO Tf< CO CO ■* 


T}< 




CM 






^H ■* O O 


^ 


03 O 


O O O O O O O 


o 




O 






o o o o 


P 


S 





































CO <N • • • • O 






a 






• • 00 • 


M C 


Ht> ■ • • . M 






K 






. . IO • 


< 


3 "3 


<# 








(N 






'. '. d ■ 


t> .... CO 




< 


J a 


CO *# CM IN lO lO CO 


C7> 


IO 


O 




CO 


CO Tjt 00 IO 


C4 


CO CN CO 00 CO CO lO 


>o 




oo 




■* 


CO 05 co ■* 


o 




OJ H Ol 00 t» N C 


00 




OO 




CO 


CO iO 00 CO 


















«*! 




* 
















61 


O O)00 O N ON 




co 


co 




iO 


(N CD <* O 


O 


15 K3 T)l H (1J lO 1- 


C 


iO 






N 


H Tf 00 N 




00 CD lO id tjh ■*' CO 


iO 


CC 


iO 




co 


od co ■* co 


s 


































O 


« v "d 
a- i ® 


OO) • • • -OS 






co 






. . OS • 

• • CO • 


( J 


&1 § 8 
O fl 2 


















O »H . . . . l- 






o 








h 
















O 


















5" a, fl a, 
6 *§* 


^ 00 O • <-H -00 






CD 




co 


IO O O C5 


!«5 

O 
H 


CN Tt< CN -CO • CN 
i-J CN -H • i-H • (N 






«> 

o 




oo 
o 


CO IO iO OJ 

d d <n « 




















CD 


•^ 
















O 


■ « OJ g 


Tj< Oi • • • • C3 






iC 




co 


. Tj* ,-H • 


PL) 

o 




^H 1-1 • • • -CO 

iH i-i • • • • O 






C 




00 

o 


•OH '. 


C) 
















OS ifl • • • • C 


"O 


O) C 




co 


• iO o • 


1 




i 


^ 


CO OS • • • -co 


<* 


CN 


l> 




OJ 




CO 

3 


3} 


■H o • • • • a 




-1 C 






.HO • 








: : !H 








ii 
o 
03 




is : 

OJ 

-fl 




H 














OJ 

3 




03 ; 

3 • 


fl G 
03 03 




a 






• OJ 

• : a P 








o 
fl 




« ! A OJ OJ 
























a 




T 


T 


T 


n~ 




g • -g O O 


















a • os _, „_ 

03 . OJ ™ ™ 

CS . H OJ CJ 

. X5 43 rfl 

d /-^ . cj cj 
C J >> 03 03 
O <13 «j qj oj 






fl fl fl >> 43 A 4= 


4= 


-c 


X 


•2 






3 


co 03 o3 03 O « C 


c 


C. 


C 








d a 3 te c3 oj « 

fl A A £ j» jv jl 
oj oj oj o 3 3 IE 
O O O £ c fl b 


a 
c 


a 
X 
P 


a 
C 


2 








^^^"^ 3 3 P 


p 


- 


P 






- aj «.fl.fl 
























cr 


e 





03 




03* . 03* 03 03 






g g g g T3 T3 X 


T 


T 


T 


T3 


X 


T3 L) "d T3 T3 






M M M W co to a 


c 
a: 


C 


C 


o 

CO 


CC 


O v-- O O O 
W CQ CO CQ 




jaquinjsj 




CN 


M •* lO © N 


00 


a 


C 






CM CO 


"* IO 



C CN 
OJ IN 
fl. OS 



•fl -S 



5o 



PULP PROCESSES AND WOOD PULPS 



419 






•J^2 ■ 



03X1 w o o 



(O CO 



r R«,2-d 






sa 
°3 



OOOOHOOOOhhhi 



oooooooooooo. 



NWQNiflCNMOONM^OO 



©©OO'-ii-HOOOOOOCO 



HOOOOHHOOHOHN 



OOHHOOHOHHHHO 



CO'*»OCOcCnOeoiO>-0''#cO»0 



0000"50005t^Ob»IN(NcD<N 



"5 £ 



1 
















>>6 rt 


fOONNMONHNCiOO 
iOOOCTHOCOiOCOOOi<Nt>-<# 


01 0) oo 




£ a 




id 


NNhhiI)OhiOO(00<0 


■*X-*C0rt<0000a><MO5CC-* 


Ah 


cocoeoi>t»«Ok050<oosoooo 




03 

s 
a 

03 

o 


COiCOC<5Tt<(Ml>00»005(NOO 




^MNiflNOSOOH^COiflH 




(oaotoiOiH(»aioo5iHH«o 




r-i |H rH i-Hi-lC* 








w 






3 


03 


t»(N©i-ia>l^i-4<N©r-<00iO 


(MlNQOOO'-iOfOl^OieOOSiO-* 








o> 


pq 


COTff~l>THiCOOOCO<OCO'-i© 




r-HCOCO COC0i-li-l<N NCOi-H 


d 












03 


OMO©NhOO«500S 






COC5>OOiOOaO<NC3<NI>OOcO 




OiOhiOOMtHINN'*tJ((00 




< 


00>O"000C0i000I>C000;Oi0O 


d ,1 


i-HCO(MC0cD>O00CCTfHO5C0Tt< 


•d d o 

a§3 


rHCOlMOOOlOOOTtHCDCOlOOCO 


<N(N<N-*i0^i-iOO<NfO>0-*00 


)— 3 "53 


IN 


o 




<D 




J 

3 


OtOiCNtOSNOOOCO© 


Or-iOH(NNHtO'*'OOOCO 


(T>t^t^C0'#Tti<CiO»C'ti-^'*00 


01 

O 


O505O5OiO>O5O5OlO3CT>OiOi»O 


>> 








o a 


^CDC5Cib-<N • rf 00 00 CO »C 




■<* <M <N GO >C CO • <N i-H CO (N <N CO 


OOOOOO 'OOOOOiO 


g M 




lid 

o> fl §0 


lOcOCOTHOOiO^-iOiNINTtHlO 


<NCDcO>-l.-iCO©COCDOiCO00a - 




hhOhhhhhhOOO^ 




c 




o3 


iN00©r^-^©iOr>.t^-*l-~© 


O 


0005Ot^lM(NOcD^HOO05CT 




COCOTttt^OOOOCOt^t-OOaSrH 








Ph 
































































o 














£ 




















» i -d 








03 o3 ? 




J3 -fl 




« « ft ft * 




O O 3 3 £ 




02 03 02 02 g 








ft 


§a 


<N CO ■* W5 S 


3 




d 










£ 





420 



CHEMISTRY OF CELLULOSE AND WOOD 



Hard Mitscherlich pulp is especially suitable for pergamyn. 
Ritter-Kellner pulp, though less suitable, can also be used for 
this purpose, as shown by the data below. A pulp for the 
manufacture of pergamyn should have a high corrected copper 
number and a low hydrate copper number. 



Lignin, 
per cent 


Copper 
number 
corrected 


7.79 


2.82 




3.31 


6.96 


2.51 




4.22 




5.16 



Ash, 
per cent 



Ritter-Kellner, hard 

Ritter-Kellner, after beating for 3 hours 

Mitscherlich, hard 

Mitscherlich, after beating for 1 hour. . 

Mitscherlich, after beating for 2.75 

hours 



0.86 
2.93 
0.95 
1.57 

1.70 



In addition to lignin and normal cellulose, all wood pulps 
contain varying amounts of pentosans and hexosans; if bleached, 
also oxycellulose. The methods and results of analyses of 
various pulps by Lenze 1 are open to question; e.g., the oxycellu- 
lose content of unbleached Mitscherlich pulp is given as 15.66 
per cent. • 

Sulphite pulp, free from lignin and ash, has the following 
composition: 2 

Per Cent 

Cellulose 87.00 

Mannan 6 . 00 

Xylan 4.50 

Lsevulan 2.5 

Specially purified pulps had only a trace of mannan, while 
normal unbleached pulps contained up to 4.18 per cent. 3 The 
mannan content of bleached Mitscherlich pulp was 0.47 per 
cent. 4 



1 F. Lenze, B. Pleus, and J. Muller, /. prakt. Chem., 101 (1920), 213-264. 

2 E. Hagglund and F. W. Klingstedt, Cellulosechemie, 5 (1924), 57. 

3 F. Lenze, B. Pleus, and J. Muller, J. prakt. Chem., 101 (1920), 213- 
264. 

4 E. Heuser and W. Dammel, Cellulosechemie, 5 (1924), 49. 



PULP PROCESSES AND WOOD PULPS 421 

Schwalbe and Becker 1 give chemical analyses of various pulps, 
from which only general conclusions can be drawn. The 
bisulphite process removes more pentosans and methoxyl than 
the alkaline processes. According to Schwalbe, 2 the pentosan 
content of a pulp is a good indication of the process by which it 
was manufactured. It would be impossible, however, to state 
from the pentosan content of a pulp that it had been made by a 
certain process without knowing the nature of the raw material. 
A sulphite pulp from a hardwood contains about the same 
amount of pentosan as a coniferous pulp made by an alkaline 
cook. 

Bleaching reduces the ash and the lignin, and increases the 
value for wood gum through the formation of oxycellulose and 
other degradation products. The reducing capacity is also 
increased accordingly. Wood pulp can be so purified by bleach- 
ing, followed by cooking with lime under pressure, that it is very 
similar to cotton in composition. 

Unbleached soda pulp is high in ash, pentosan, and lignin, and 
low in resin, "wood gum," and reducing substances. Un- 
bleached Mitscherlich pulp is high in ash, resin, lignin, and 
pentosan, and has a relatively high reducing property; in com- 
parison, Ritter-Kellner pulp has correspondingly lower values. 

The strength of kraft pulp, which is made from coniferous 
woods, is due to the length of the fibers and to mild cooking 
conditions. The reactions are not sufficiently drastic to remove 
all of the incrustants, so that the cellulose is not so subject to 
attack. In the Mitscherlich process, the pulp is washed in situ 
and removed from the digester by hand. This procedure pre- 
serves the length of the fibers, which is decreased by blowing the 
digester. The strength of kraft pulp appears to be due in part 
also to the presence of resins and fats, which serve as binding 
agents. 3 The glyceride of linoleic acid is of particular impor- 
tance, since it is gradually converted into linoxyn by oxidation. 
The best pulp is obtained from green wood, since in the old, the 
linoxyn is already largely formed. 

1 C. G. Schwalbe and E. Becker, Paper 27, 25 (1921), 20. 

2 C. G. Schwalbe, Z. angew. Chem., 31 (1918), 52. 

3 C. G. Schwalbe, Papier-Fabr. Fest u. Auslandsheft, 21 (1923), 60; 
A. Abadie, Paper Trade J., 51, 6 (1910), 46. 



422 



CHEMISTRY OF CELLULOSE AND WOOD 



The resin content of various pulps is shown in the following 
table: 1 



Ether 
extract 



Alcohol 
extract 



Total 
extract 



Spruce wood 

Kraft 

Soda, unbleached 

Soda, from aspen 

Soda, bleached 

Mitscherlich, unbleached . . 
Mitscherlich, bleached 
Ritter-Kellner, unbleached 
Ritter-Kellner, bleached . . . 

Nitrating pulp 

Nitrating cotton 

Pure cotton 



0.60 
0.17 
0.22 
0.59 
0.05 
0.68 
0.70 
0.86 
0.65 
0.16 
0.14 
0.10 



0.38 
0.43 
0.20 
0.76 
0.24 
0.31 
0.52 
0.49 
0.45 
0.35 
0.23 
0.20 



0.98 
0.60 
0.39 
1.35 
0.29 
1.01 
1.21 
1.35 
1.11 
0.51 
0.37 
0.30 



The resin from sulphite pulp gives the cholesterol reaction with 
acetic anhydride and sulphuric acid, while that from soda and 
sulphate pulps does not. 

Hagglund 2 was unable to find any consistent relationship 
between the mechanical properties of pulps and their chemical 
constants. It sometimes happens that a well-cooked pulp is 
as strong or stronger than an undercooked pulp. 

The hygroscopic moisture varies with the purity of the pulp, 
the temperature, and the humidity of the atmosphere. Lespius 3 
determined the hygroscopic moisture in various pulps at a 
temperature of 20° and at a humidity of 55 per cent. The 
results in the following table are based on the dry weight of the 
fibers : 

Hygroscopic 

Moisture, 
Per Cent 

1. Aspen, white mechanical pulp 11.7 

2. Spruce, white mechanical pulp 1 1 . 94 

3. Spruce, brown mechanical pulp 11 .71 

4. Straw soda pulp, bleached 9 . 25 

5. Wood soda pulp, unbleached 9 . 40 

6. Wood soda pulp, bleached 9 . 80 

7. Wood sulphite pulp, unbleached 10 . 07 

1 C. G. Schwalbe, Z. angew. Chem., 31 (1918), 58. 

2 E. Hagglund, Papier-Fabr., 17 (1919), 301. 

3 B. Lespius, Ber., 18 (1885), 2491. 



PULP PROCESSES AND WOOD PULPS 423 

Certain incrustants such as occur in Mitscherlich pulps have a 
high hygroscopicity. 1 Mitscherlich pulps for pergamyn manu- 
facture are more hygroscopic than cellulose, while Ritter-Kellner 
pulps are less so. 

Discoloration of Sulphite Pulp. — Sulphite pulp on storage, 
especially in moist air, frequently becomes pink or brown. 
Muth 2 thought that the discoloration was due to decomposition 
of tannin in the wood during boiling with sulphite liquor. The 
same opinion was held by Harpf. 3 Unbleached pulp treated 
with sodium hydroxide, lime water, and nitric acid gave the 
same coloration as waste sulphite liquor. Ferric chloride gave 
a green color with wood and a violet color with sulphite pulp, 
but no characteristic reaction for tannin could be obtained with 
the spent liquor. He found that oxidizing agents developed the 
color and that this could be removed by the use of antichlors 
or fresh sulphite liquor. 

According to Klemm, 4 the coloration is due to reduction of 
the tannin by sulphur dioxide and its subsequent oxidation. 
Support for this view was obtained by heating tannin with sul- 
phurous acid, whereby a light-colored solution turning red with 
oxidizing agents was obtained. On dipping paper into this solu- 
tion and exposing it to light and air a red coloration developed. 

Legrand 5 attributes the coloration to oxidation of residual 
incrusting substances. The tendency to color is greater the 
lower the purity of the pulp. Mitscherlich pulp is more suscepti- 
ble to staining than Ritter-Kellner. The color is apparently 
not due to tannin but to the oxidation of residual lignin or lignin- 
sulphonic acid. 6 Freshly precipitated ligninsulphonic acids 
show the same coloration with oxidizing agents as unbleached 
pulp. The substance actually producing the coloration has 
not been isolated. A specific preventative agent was found in 
potassium persulphate. The pulp is agitated with 0.5 per cent 
of potassium persulphate and an equal weight of sulphuric acid 

1 C. G. Schwalbe and E. Becker, Paper, 27, 25 (1921), 20. 

2 E. Muth, Dinglers poly tech. J., 291 (1894), 235. 

3 A. Harpf, Wochbl. Papierfabr., 26 (1895), 586, 641. 

4 P. Klemm, Wochbl. Papierfabr., 40 (1909), 2882. 

5 Legrand, Papier-Fabr., 9 (1910), 32-33. 

6 E. Heuser and S. Samuelsen, Papier-Fabr., 20 (1922), 1249, 1285, 
1321, 22 (1924), 130. 



424 CHEMISTRY OF CELLULOSE AND WOOD 

or aluminum sulphate for 8 to 9 hours. The exact nature of the 
reaction is unknown. The pulp so treated contains the same 
amount of lignin as the original and is not bleached. Evidently, 
the lignin is so modified that it is incapable of subsequent 
reddening. 

Copper and especially iron, which is usually derived from the 
limestone, catalyze the oxidation of the impurities and promote 
the coloration. 1 Impure alum is also effective. 2 Prevention 
of coloration is obtained by treating the pulp with sulphurous 
acid or disodium phosphate. 

Bleaching. — The bleaching of sulphite pulp with sodium 
hypochlorite takes place in four stages. 3 In the first stage the 
consumption of chlorine is very rapid. If the solution is acid, a 
red compound which is rapidly oxidized is first formed. This 
coloration is not obtained in an alkaline solution. During the 
second stage the yellow impurities are gradually oxidized, while 
in the third there is a slow oxidation of the colored and colorless 
oxidation products; unless these are completely removed, the 
pulp becomes yellow on storage. In the last stage, the cellulose 
itself is attacked, particularly in an acid medium. 

The rate of bleaching is slow in alkaline solutions, and in 
acid solutions increases in proportion to the square of the acidity. 3 
A rise of 7° doubles the rate of bleaching. Between wide limits, 
the rate of bleaching is independent of the concentration of the 
hypochlorite. Samples bleached at the higher temperatures are 
whitest on the basis of equal consumption of chlorine. Owing 
to decomposition of the hypochlorite and oxidation of the 
cellulose, greater quantities of chlorine are required to produce 
the same whiteness, the higher the acidity. 

The bleaching action at 38° was eight times more rapid than 
at the ordinary temperature. 4 A higher temperature caused 
weakening of the fiber. Pulp that is being circulated can be 
bleached with 40 per cent of the amount of chlorine required 
for pulp not kept in motion. The cause is incorrectly attributed 
to the bleaching effect of atmospheric oxygen. 

1 H. E. Wahlberg, Papier-Fabr., 21 (1923),204-205; 22 (1924), 229-230. 

2 0. L. Berger, Paper, 30, 5 (1922), 16. 

3 J. Nussbaum and W. Ebert, J. Soc. Chem. Ind., 26 (1907), 1063. 

*C. Beadle and H. P. Stevens, J. Soc. Chem. Ind., 33 (1914), 727-730. 



PULP PROCESSES AND WOOD PULPS 



425 



The rate of bleach consumption is a linear function of the 
logarithm of the time during the main course of the reaction and 
(probably) within the temperature limits of to 50°. The final 
color of the pulp is largely independent of the bleaching tem- 
perature. 1 The effect of temperature on the time of bleach 
consumption is shown in the following table: 

Time Required for 90 Per Cent Bleach Consumption 



Temperature, 
degrees Centigrade 


Total time, 
minutes 


Point of 90 

per cent bleach 

consumption, 

minutes 


Per cent of total 

time required to 

reach 90 per cent 

bleach consumption 


21 
35 
41 
46 


713 
220 
130 

84 


220 
75 
50 
34 


30.9 
34.2 
38.5 
40.5 



The formation of yellow, stable chloramines 2 does not appear 
to be a factor in the bleaching of wood pulps, owing to the small 
amount of nitrogenous compounds present. Hard water is 
without effect upon the rate of bleaching, color, and yield of 
bleached pulp. 3 The final color of the pulp is independent of 
consistencies up to 7 per cent with temperatures ranging from 
21 to 46°. The rate of bleach consumption increases with the 
consistency, this effect being greatest at lower temperatures. 4 

Pulp is ordinarily bleached with a solution of bleaching 
powder, in its normally alkaline condition, at temperatures of 30 
to 40°. An alkaline reaction is superior to an acid one. The 
best results are obtained by bleaching at the ordinary tempera- 
ture with two-thirds of the bleach liquor acidified with carbon 
dioxide, then completing the operation at 30 to 35° with the 
remaining one- third of ordinary alkaline bleach liquor. 5 

Bleaching experiments conducted in a ball mill at 30 to 35° 
gave a better yield and quality of pulp in a shorter time than cold 

1 C. E. Curran and P. K. Baird, Paper Trade J., 79, 11 (1924), 45; G. K. 
Spence, Tech. Ass. Pulp and Paper Ind., Series 4 (1921), 39. 

2 C. F. Cross, E. J. Bevan, and J. F. Briggs, J. Soc. Chem. Ind., 27 
(1908), 260; Chem. Ztg., 32 (1908), 489; R. E. Crowther, J. Soc. Dyers 
Colourists, 29 (1913), 38. 

3 C. E. Curran and P. K. Baird, Paper Trade J., 79, 3 (1924), 41-43. 

4 C. E. Curran and P. K. Baird, Paper Trade J., 80, 16 (1925), 51-53. 

5 V. Hottenroth, Zellstoff u. Papier, 2 (1922), 6-12. 



426 CHEMISTRY OF CELLULOSE AND WOOD 

bleaching. 1 Passage of a current of air through the apparatus 
cut the time in half for warm bleaching, but had no influence in 
the cold. The a-cellulose value was reduced by temperatures 
above 35° and by prolonged cold bleaching. The effect of 
air is purely mechanical, since nitrogen had the same effect. 2 
Carbon dioxide is formed by the action of hypochlorite solutions 
on all forms of cellulose. For every gram of active chlorine 
consumed per 100 grams of cellulose, 0.26 to 0.28 gram of carbon 
dioxide is produced. This forms calcium bicarbonate with the 
lime in the bleaching powder. The calcium bicarbonate has a 
buffer action on the hydrogen-ion concentration, reducing it to a 
point where the bleaching action comes to a standstill. Removal 
of the carbon dioxide and dissociation of the bicarbonate are 
facilitated by agitation, heat, and a current of air. 

The normal chlorine consumption may be reduced 45 to 54 
per cent if the pulp is treated with chlorine gas, washed with 
water or preferably sodium hydroxide, and the bleaching com- 
pleted with bleaching powder. 3 Wells 4 had shown that the 
bleach consumption could be reduced one-half by conducting 
the bleaching operation in two stages, with intermediate washing. 

Moderate bleaching, as shown by the preparation of test 
papers, increased the strength of the pulp. 5 According to Froh- 
berg, 6 normal bleaching produces decided loss of strength. The 
folding test may be reduced 42 per cent and the tensile strength 
11 per cent. Overbleaching with 2 per cent excess of bleaching 
powder gives a soft opaque fiber resembling cotton. 

Permanganates are more efficient bleaching agents than 
bleaching powder. One part of oxygen in the form of potassium 
permanganate was as effective as 5.5 parts of oxygen in the form 
of bleaching powder. 7 According to Sutermeister, 8 one part 

1 C. G. Schwalbe and H. Wenzl, Papier-Fabr., 20 (1922), 1625-1631. 

2 C. G. Schwalbe and H. Wenzl, Papier-Fabr., 21 (1923), 268, 385; 
Z. angew. Chem., 36 (1923), 302-304. 

3 E. Heuser and W. Niethammer, Papier-Fabr. Fest u. Auslandsheft, 
21 (1923), 52-60. 

4 S. D. Wells, Paper Trade J., 71, 22 (1920), 34; /. Ind. Eng. Chem., 13 
(1921), 936. 

5 E. Sutermeister, Papier-Fabr., 12 (1914), 898. 

6 A. Frohberg, Wochbl. Papierfabr., 44 (1913), 3599. 

7 C. Beadle and H. P. Stevens, /. Soc. Chem. Ind., 33 (1914), 727. 

8 E. Sutermeister, I.e. 



PULP PROCESSES AND WOOD PULPS 



427 



of potassium permanganate produced the same effect as 1.85 
parts of calcium hypochlorite; hence the oxygen ratio would be 
1:3.6. 

The discoloration of bleached pulps may be due to residual 
chlorinated and oxidized lignin derivatives, and to oxy cellulose. 
Bachelder 1 heated samples of pulp at 30, 70, 75, 87, and 102°, 
for 16 hours. Unbleached sulphite and sulphate pulps showed 
little change of color, while pronounced colorations were obtained 
with unbleached soda pulp from aspen, and bleached sulphite 
pulps. 

According to Cross and Bevan, 2 the non-cellulose constituents 
are first chlorinated, while beyond a certain point there is hydrol- 
ysis and oxidation of the cellulose. The contention that the 
cellulose molecule is chlorinated should be accepted with reserve. 

The inorganic and organic chlorine present in bleached sulphite 
pulps has been determined by Schwalbe. 3 The discoloration 
of bleached pulps is attributed to oxycellulose rather than to 
organic chlorine compounds, as held by Cross and Bevan. 
Chlorine Content of Bleached Pulps 



Bleaching agent 



Ash, 
per 
cent 



Chlorine 
in ash, 
per cent 



Organic 
chlorine, 
per cent 



Total 
chlorine, 
per cent 



Ca(OCl 2 ) 

Ca(OCl) 2 , double quantity 

NaOCl 

NaOCl, double quantity 

Ca(OCl) 2 , acid 

Ca(OCl) 2 , acid, double quantity 

NaOCl, acid 

NaOCl, acid, double quantity. . 

Unbleached pulp 

Commercial bleached pulp 



0.83 


0.011 


0.058 


1.35 


0.014 


0.053 


0.47 


0.012 


0.071 


0.80 


0.055 


0.059 


0.36 


0.014 


0.052 


0.48 


0.008 


0.067 


0.37 


0.002 


0.049 


0.68 


0.006 


0.047 


0.60 




0.012 


0.99 


0.011 


0.042 



0.069 
0.068 
0.083 
0.114 
0.066 
0.075 
0.051 
0.053 
0.012 
0.053 



The writer has examined bleached sulphite pulp, purified by a 
special process, that contained oxycellulose yet did not turn 
yellow on prolonged heating at 70°. 

1 C. L. Bachelder, Paper, 24 (1919), 1096-1097. 

2 C. F. Cross and E. J. Bevan, J. Soc. Chem. Ind., 9 (1890), 450; Chem. 
Ztg., 32 (1908) 489. 

3 C. G. Schwalbe, Z. angew. Chem., 21(1908), 302-303. 



CHAPTER XIII 
THE DISTILLATION OF CELLULOSE AND WOOD 

Cellulose is profoundly changed by prolonged heating at 
moderate temperatures. Cotton 1 placed over steam pipes for 
some months carbonizes completely at a temperature of 120 to 
130°. Knecht 2 observed that bleached cotton exposed on watch 
glasses at a temperature of 93° turned grayish brown after a 
few days with the formation of oxy cellulose. Heated in sealed 
tubes, the cotton took up oxygen and decreased 50 per cent in 
strength. 

Cellulose heated with water at 340° for about 12 hours is 
changed, with the evolution of carbon dioxide, into a substance 
resembling soft coal. 3 The reaction takes place approximately 
as follows : 

4C 6 H 10 O 5 = C 2 iH 16 2 + 3C0 2 + 12H 2 0. 

If cellulose is heated with benzene under pressure at 250 to 260°, 
it gradually decomposes, but with only a slight evolution of gas, 
to dark-brown, tar-like products, soluble in the benzene. 4 

Destructive Distillation of Cellulose. — Subjected to destructive 
distillation, cellulose shows an exothermic reaction, though this 
is not marked if the distillation is sufficiently slow. Chorley and 
Ramsay 5 claimed an exothermic reaction with raw cotton but 
not with the purified product (absorbent cotton); however, the 
data show with the latter a rise of 23° within the retort compared 
with 54° for jute. The purest cotton obtainable has been 

1 C. Koechlin, Bull. Mulhouse, 58 (1888), 547. 

2 E. Knecht, J. Soc. Dyers Colourists, 36 (1920), 195-198. 

3 F. Bergius, J. Soc. Chem. Ind., 32 (1913), 462; S. Stein, Chem. Zentr., 
72, II (1901), 950. 

4 F. Fischer and W. Schneider, C. A., 14 (1920), 2081. 

5 J. C. Chorley and W. Ramsay, J. Soc. Chem. Ind., 11 (1892), 873. 

428 



THE DISTILLATION OF CELLULOSE AND WOOD 



429 



observed 1 to give a marked and sudden evolution of heat 
at 345°. 

The evolution of water from heated cellulose is most marked 
at about 290° and the exothermic reaction would be anticipated 
at this point. In fact, Bantlin 2 found that with cellulose the 
exothermic reaction took place between 250 and 300°. On 
heating cellulose up to 1000° over a period of 80 hours, the evolu- 
tion of water took place as follows: 3 



Temperature, 

degrees 

Centigrade 


Water, 
per cent 


Temperature, 

degrees 

Centigrade 


Water, 
per cent 


190 
240 
290 
350 


0.43 

1.57 

22.61 

6.57 


420 

510 

600 

1,000 

Total 


2.42 
1.94 
1.03 . 
0.0 




36.57 



Only one-half as much water is obtained by rapid as by slow 
distillation. 

Distillations of cotton and wood celluloses, prepared by the 
sulphite process, have been made by Klason 4 and his associates, 
with the following results: 



1 H. Hollings and J. W. Cobb, J. Chem. Soc, 107 (1915), 1110. 

2 G. Bantlin, J. Gasbel, 57 (1914), 32, 55. 

3 O. Rau and G. Lambris, /. Gasbel, 56 (1913), 561. 

4 P. Klason, G. von Heidenstam, and E. Norlin, Z. angew. Chem., 22 
(1909), 1205-1214. 



430 



CHEMISTRY OF CELLULOSE AND WOOD 



Cellulose from 



Cot- 
ton, 
per 

cent 



Pinus 






Picea 


silves- 






excelsa, 


tris, 






per 


per 


cent 


cent 





Betula 
alba, 
per 
cent 



Fagus 

silva- 

tica, 

per 

cent 



Composition of raw material: 

Water 

Ash 

Organic matter 

Distillation products: 

Charcoal 

Carbon dioxide 

Ethylene 

Carbon monoxide 

Methane 

Methyl alcohol 

Acetone 

Acetic acid 

Organic matter in 

CH 3 .COONa 

Tar 

Water 

Loss 



4.82 

0.13 

95.05 

38.82 

10.35 

0.17 

4.15 

0.27 



0.07 
1.39 

5.14 

4.18 

34.52 

0.94 



5.60 


6.44 


0.44 


0.49 


93.96 


93.07 


36.93 


34.86 


12.83 


11.94 


0.21 


0.19 


3.40 


3.92 


0.27 


0.22 


Trace 


0.07 


0.08 


0.13 


2.18 


2.79 


4.22 


8.50 


4.85 


6.28 


34.17 


29.99 


0.86 


1.11 



6.65 

0.92 

92.43 

33.39 

11.14 

0.41 

3.49 

0.47 

0.15 

3.89 

7.72 

9.58 

29.35 

0.40 



7.81 

0.77 

91.42 

32.91 
11.96 
0.25 
3.80 
0.39 
0.19 
0.26 
3.50 

8.67 

5.23 

31.88 

0.93 



The decomposition of the various celluloses is quite similar. 
The small amounts of methyl alcohol obtained from the wood 
celluloses, without doubt, had their origin in residual methoxyl 
groups. Cotton does not give methyl alcohol, though the 
presence of 0.277 per cent of methoxyl in absorbent cotton has 
been reported. 1 The wood celluloses, particularly hardwood, 
gave markedly greater yields of acetic acid than cotton. This 
may be due to the higher hemicellulose content; Bergstrom, 2 





Cellulose 


Starch 


Sugar 


Acetic acid, per cent 


3.28 


5.29 


8.78 



1 L. F. Hawley and S. S. Aiyar, /. Ind. Eng. Chem., 14 (1922), 1055. 
2 H. Bergstrom, Papier-Fabr., 11 (1913), 759. 



THE DISTILLATION OF CELLULOSE AND WOOD 431 

however, obtained only 0.29 per cent of acetic acid by the distil- 
lation of birch "wood gum." The results obtained by Bantlin 1 
on distilling cellulose, starch, and sugar indicate that the yield 
of acetic acid decreases with the increase in molecular complexity 
of the carbohydrate. 

The distillate obtained from filter paper has been examined 
by Erdmann and Schaefer. 2 They used a kilogram of paper per 
charge and completed the distillation in 2 hours. The gas 
contained by volume: 65.5 per cent of carbon monoxide; 19.0 
per cent of methane; 11.5 per cent of hydrogen; 2.4 per cent of 
nitrogen; 0.9 per cent of oxygen; 0.5 per cent of hydrocarbons; 
and only 0.2 per cent of carbon dioxide. The aqueous distillate 
contained acetone, formaldehyde, furfural, hydroxymethyl- 
furfural, maltol, and 7-valerolactone. 

Filter paper distilled with superheated steam gave 2.5 to 2.8 
per cent of acetic acid but no methyl alcohol. 3 

Cotton distilled with or without zinc gave acetone, methyl- 
ethyl ketone, furan, a-methylfuran, 2,5-dimethylfuran, 
trimethylfuran, tetramethylfuran, phenol, cresol, acetic 
acid, traces of toluene, and a compound, C12H14O, possibly 
hydrodiphenyleneoxide. 4 

Diacetyl 5 is also formed. Phenol was the only phenolic con- 
stituent found by Wichelhaus 6 in the distillate from 20 kilograms 
of pure cotton. 

Maltol has not been reported as present in wood distillates, 
though Cross and Bevan 7 obtained it by the incipient roasting 
of cellulose and lignocellulose, and suggest that the pyrone 
ring may be represented in the constitutional formula of cellu- 
lose. Maltol is a crystalline compound, m.p. 159°, readily 
soluble in water and giving a violet color with ferric chloride. 
It is formed in the roasting of malt, 8 and occurs naturally in 



1 G. Bantlin, I.e. 

2 E. Erdmann and C. Schaefer, Ber., 43 (1910), 2398. 

3 G. Buttner and H. Wislicenus, J. prakt. Chem., 79 (1909), 177. 

4 J. Sarasin, Arch. sci. phys. nat., [IV] 46 (1918), 5. 
b J. M. Johlin, J. Am. Chem. Soc, 37 (1915), 892. 

6 H. Wichelhaus, Ber., 43 (1910), 2922. 

7 C. F. Cross and E. J. Bevan, J. Soc. Dyers Colourists, 32 (1916), 135. 

8 J. Brand, Ber., 27 (1894), 806; H. Kiliani and M. Bazlen, Ibid., 3115. 



432 CHEMISTRY OF CELLULOSE AND WOOD 

the needles of white fir (Abies alba Mill.) 1 and larch bark. 2 It 
is identical with the larixic acid of Stenhouse. 3 An isomaltol, 4 
m.p. 98°, has been isolated from bread and various roasted food 
products. It is probably /3-methyl-/3 : hydroxy-7-pyrone, while 
maltol is a-me thy l-^-hydroxy-y-py rone. 5 

HC — O — CH HC — — C.CH 3 

II II II II 

CH 3 .C— CO— C.OH HC— CO— C.OH 

Isomaltol . Maltol 

Vacuum Distillation of Cellulose. — In 1918, Pictet and Sarasin 6 
reported the interesting discovery that when cellulose or starch 
is distilled in vacuo, Z-glucosan is obtained in a yield of 45 per cent. 
At a pressure of 12 to 15 millimeters, a thick yellow oil is obtained, 
from which Z-glucosan can be crystallized by the use of acetone; 
white crystals, m.p. 179.5°; [a] D = —66.25°; triacetate, m.p. 
110°; and tribenzoate, m.p. 199.5°. Laevoglucosan has the 
formula C 6 Hi O 5 , contains three hydroxyl groups, and is an 
anhydride of glucose. Complete hydrolysis with dilute acids 
at normal pressure requires about 6 hours, and glucose is the 
only sugar obtained. 

Since cellulose gives furan derivatives on destructive distilla- 
tion, and especially since Z-glucosan gave co-brommethylfurfural 
with hydrobromic acid, Pictet and Sarasin 1 concluded that 
both cellulose and Z-glucosan contain the furan ring, and that 
cellulose is a polymeride of Z-glucosan. It was then observed 
that, while ordinary glucose (a-glucose) gave only traces of 
Z-glucosan, the latter could be obtained in quantity from many 
natural /3-glucosides; 7 also that when glucose is heated at 150 
to 155° at a pressure of 15 millimeters glucosan (a-glucosan) is 

1 W. Feurstein, Ber., 34 (1901), 1804. 

2 A. Peratoner and A. Tamburello, Ber., 36 (1903), 3407. 

3 J. Stenhouse, Ann., 123 (1862), 191. 

4 A. Backe, Compt. rend., 150 (1910), 78, 540. 

5 A. Peratoner and A. Tamburello, Chem. Zentr., 76, II (1905), 680. 

6 A. Pictet and J. Sarasin, Compt. rend., 166 (1918), 38-39; Helvetica 
Chim. Acta, 1 (1918), 87-96; J. Sarasin, Arch. sci. phys. not., [IV] 46 (1918), 
5-32. 

7 A. Pictet and H. Goudet, Helvetica Chim. Acta, 2 (1919), 698. 



THE DISTILLATION OF CELLULOSE AND WOOD 433 

obtained. 1 Glucosan has the m.p. 108 to 109°, and is d-rotatory ; 
[a] D = +69.8°. 

With the idea that the configuration of the glucose might have 
a bearing on the glucosan obtained, Karrer 2 distilled j8-glucose 
and, in fact, obtained almost the same yield of /-glucosan as 
from starch. This observation strongly indicated that the 
0-glucosidal linkage exists in starch and cellulose and that 
/-glucosan and glucosan are formed from (3- and a-glucose 
respectively. 

Pictet 3 then derived the two glucosans by dehydration of 
Boeseken's 4 formulae for a- and /3-glucose. Pictet's formula 



h— c— OH 

H-C-OH 

1 ° 

HO— C— H 

H-i-J 

I 
H-C-OH 

CH2OH 

a-Glucose 



I 
HO— C— H 

I 
H— C— OH 
I o 

HO— C— H 

I 
H— C— OH 

CH2OH 
^-Glucose 



I 
H— C\ 

l> 
H— (X 



HO 



i-H° 

H— C— OH 

CH2OH 

Glucosan 
a-Glucosan) 



I 
H— C — 

I 
H— C 

I O 

HO— C— II I 

H-t- 

II- 



-C— OH 

I 
CH2 — 

(a-Glucosan) 
{Irvine) 



■i-H 



H-C-OH 



HO— C— H 

I 
H— C — 

II— C— OH 



CHi 



Lsevoglucosan 
(/3-Glucosan) 



for /-glucosan has been confirmed by Irvine and Oldham, 5 who 
proved it to be 1, 6-0-glucose anhydride. A different structure 
was suggested by them for glucosan, though the correctness of 
the Pictet formula is rendered highly probable by the work of 
Cramer and Cox. 6 

Irvine and Oldham 5 showed conclusively that there was no 
structural relationship between /-glucosan and cellulose, since 
they yield entirely different trimethylglucoses. They suggest 
the expressions a-glucosan and /3-glucosan, though the term 



1 A. Pictet and P. Castan, Compt. rend., 171 (1920), 243; Helvetica Chim. 
Acta, 3 (1920), 645. 

2 P. Karrer, Helvetica Chim. Acta., 3 (1920), 258. 

3 A. Pictet, Helvetica Chim. Acta, 3 (1920), 649; A. Pictet and M. 
Cramer, Ibid., 640. 

4 J. Boeseken, Ber., 46 (1913), 2612. 

5 J. C. Irvine and J. W. H. Oldham, /. Chem. Soc, 119 (1921), 1744-1759. 

6 M. Cramer and E. H. Cox, Helvetica Chim. Acta., 5 (1922), 884. 



434 CHEMISTRY OF CELLULOSE AND WOOD 

0-glucosan had been previously substituted for /-glucosan by 
Vongerichten and Muller. 1 

When heated in a vacuum and under pressure, especially 
in the presence of zinc chloride and platinum black, both glucosan 
and /-glucosan undergo various degrees of polymerization. 
Di-, tetra-, hexa-, and octa-/-glucosans, 2 and di- and tetra- 
glucosans, 3 have been obtained. Tetra-a- and tetra-0-glucosans 
are easily and completely acetylated or methylated. 4 

Distillation of ethylcellulose in vacuo gave only a small amount 
of oil and no derivative of glucosan. 5 Reilly, 6 working with 
methylcelluloses containing 25.3 to 43.7 per cent methoxyl, 
obtained a product corresponding to dimethylglucosan from a 
methylcellulose containing 28.1 per cent methoxyl. It gave a 
dimethylglucose on hydrolysis. 

A high yield of /-glucosan is dependent upon a pure cellulose 
and rapid distillation. The removal from cotton of compounds 
soluble in water, alkali, and acid greatly improves the yield. 
The highest yield obtained by Venn 7 from cotton was 37 per 
cent, while 40 per cent was obtained from hydrocellulose. Irvine 
and Oldham 8 distilled 4300 grams of starch, obtaining 1670 
grams of syrup, from which 230 grams of pure /-glucosan was 
obtained. During the distillation, acids are formed that hydro- 
lyze the starch and cellulose to /3-glucose, which dehydrates and 
passes over as /-glucosan. 

In connection with the occurrence of maltol in fir needles and 
larch bark, it is of interest that /-glucosan was first obtained 
from the glucosides of conifers. 9 

Destructive Distillation of Wood. — The oldest process by 
which wood was transformed into products of economic value by 
chemical reactions, is wood distillation. This is not surprising, 
since a knowledge of the combustibility of wood must have been 

1 E. Vongerichten and F. Muller, Ber., 39 (1906), 241. 

2 A. Pictet and J. H. Ross, Helvetica Chim. Acta, 5 (1922), 876-883. 

3 A. and J. Pictet, Helvetica Chim. Acta, 4 (1921), 788-795. 

4 H. Pringsheim and K. Schmalz, Ber., 55 (1922), 3001-3007. 

5 K. Hess and W. Wittelsbach, Z. Elektrochem., 26 (1920), 237. 

6 J. Reilly, Helvetica Chim. Acta, 4 (1921), 616-621. 

7 H. J. P. Venn, /. Textile Inst, 15 (1924), 414. 

8 J. C. Irvine and J. W. H. Oldham, I.e. 

9 C. Tanret, Compt. rend., 119 (1894), 158. 



THE DISTILLATION OF CELLULOSE AND WOOD 435 

forced upon the attention of prehistoric man. There naturally 
followed in the course of ages the observation that limited com- 
bustion produced tar and pyroligneous acid in addition to 
charcoal. 

The distillation industry uses both hardwoods and softwoods, 
the most suitable species available being selected. The hard- 
woods are especially valuable for their high yields of acetic acid 
and methyl alcohol. Softwoods, on the other hand, give low 
yields of acetic acid and methyl alcohol, but produce turpentine 
and tars of special value ; hence, unless the charcoal is of primary 
importance, only the most resinous species are utilized. 

Hardwood is usually distilled by heating it in the form of 
cord wood or short blocks in closed retorts. The distillate 
settles into two layers, the upper consisting of pyroligneous acid 
and the lower of tar. This settled tar contains some acetic 
acid and methyl alcohol that are recovered by distilling the tar 
with steam. The pyroligneous acid contains the "soluble tar." 
This is separated as a residue by distillation, the distillate con- 
sisting essentially of acetic acid and methyl alcohol. After 
neutralizing the distillate with lime, the solution is distilled to 
remove the methyl alcohol, calcium acetate remaining in the still. 

Longleaf and Cuban pines are the only native conifers dis- 
tilled on a large scale. The procedure is similar to hardwood 
distillation. The methyl alcohol and acetic acid are not 
recovered, except under special conditions. The tar, owing to 
the presence of terpenes and decomposition products of resin, 
floats on the pyroligneous acid. The crude tar is refined by 
distillation and treatment with chemicals into "wood turpen- 
tine," pine oil, tar, tar oils, and pitch. Pitch is obtained directly 
from the retort towards the end of the distillation and as a 
residue from the distillation of the tar. 

The extraction process as applied to resinous woods involves 
no chemical reactions. The comminuted wood is distilled with 
steam to remove the turpentine and pine oil, and then digested 
with a special gasoline to remove the rosin. The solvent is 
recovered by distillation and reused. 

The data obtained in the distillation of wood should always 
be interpreted in the light of the experimental conditions as the 
results may vary greatly according to : 



436 CHEMISTRY OF CELLULOSE AND WOOD 

1. Species, dimensions, and moisture content of the wood. 

2. Speed and temperature during distillation. 

3. Method of heating. 

The first systematic investigations on wood distillation were 
conducted by Violette. 1 Wood, in the form of cylinders 6 centi- 
meters long and 1 centimeter in diameter, was treated with 
superheated steam, the distillation temperature being measured 
within the retort near the egress of the steam and distillation 
products. Owing to the uniform heating of the charge, the 
yield of charcoal was greatly increased. Violette was primarily 
interested in obtaining charcoal suitable for gunpowder, hence 
little attention was paid to the distillate. He found that carbon- 
ization took place at about 250°; at 300° "red" charcoal, and at 
350° black charcoal were formed. His results should be distin- 
guished from those obtained under industrial conditions where 
most of the heat is supplied externally. 

Under ordinary conditions of distillation, the charcoal retains 
the structure of the wood. Violette 2 observed that when heated 
to 300 to 400° in an entirely closed vessel, the wood seemed to 
melt and flow. When cold it formed a structureless, black, 
glistening mass, like tar, which contained 67 per cent of carbon and 
27 per cent of oxygen. According to Petzholdt, 3 the product 
resembles coal. 

It has been observed that wood becomes very plastic under 
pressure when heated at 200 to 250°, temperatures below the 
carbonization point. 

Exothermic Reaction. — The distillation of wood is accom- 
panied by a violent exothermic reaction at a temperature of 
about 275°, as was first noted by Chorley and Ramsay. 4 The 
increase in temperature due to the exothermic reaction was 30° 
with oak, 55° with beech, and 40° with alder. According to 

1 H. Violette, Compt. rend., 26 (1848), 683-685; 27 (1848), 53-55, 55-56; 
32 (1851), 713-717; 38 (1854), 107-116; Ann. chim. phys., [3] 23 (1848), 
475-508; [3] 32 (1851), 304-350; [3] 39 (1853), 291-342. 

2 H. Violette, Compt. rend., 32 (1851), 715; Ann. chim. phys., [3] 32 
(1851), 326. 

3 A. Petzholdt, "Beitrag zur Kenntnis der Steinkohlenbildung, " Leipzig 
(1882), p. 11; cited by E. Donath, Chem. Ztg., 32 (1908), 1272. 

4 J. C. Chorley and W. Ramsay, J. Soc. Chem. Ind., 11 (1892), 395-403, 
872-874. 



THE DISTILLATION OF CELLULOSE AND WOOD 437 



Hornsey 1 , when the heat is withdrawn, the exothermic reaction 
produces a rise of 83° with chestnut and 136° with pine wood. 2 

The thermochemistry of wood distillation has been fully inves- 
tigated by Klason. 3 Distillation begins at 250°, becomes rapid 
at about 275°, and is completed at 350°. The following table 
gives the data obtained for birch wood. 4 

Thermochemistry of the Distillation of Birch Wood for 8 Hours 
at Ordinary Pressure 



Products 



Per cent 
by weight 



Heat of combustion 
Calories 



Per 

kilogram 



On per 
cent of 

product 



Heat 
absorbed 
by prod- 
ucts at 

275° 



Charcoal 

Tar 

Acetic acid 

Formic acid 

Methyl alcohol 

Acetone 

Formaldehyde 

Carbon dioxide 

Carbon monoxide 

Methane 

Ethylene 

Volatile oils 

Organic matter, undetermined 
Water 

Sum 

Birch wood 



30.85 

16.94 
6.77 
0.61 
1.49 
0.20 
1.00 

10.17 
3.57 
0.98 
0.25 
3.00 
3.69 

20.48 

100.00 



7,860 


2 


,428.3 


6,053 


1 


,025 


3 , 500 




237 


1,341 




8.2 


5,310 




79.1 


7,300 




14.6 


4,566 




45.7 


2,430 


86.8 


13,060 




128 


11,860 




29.6 


7,875 




236.3 


8,598 




317.3 




4 


,636.4 


4,895 





20.1 

40.2 

14.3 

1.4 

5.8 

1.1 

5.6 

6.1 

2.4 

1.8 

0.3 

7.1 

8.8 

147.5 

262.5 

100.6 



All results are calculated for 1 kilogram of wood at the 
exothermic temperature, 275°. Since the heat of combustion 
of birch wood is 4895 Calories, and 100.6 Calories are required 

1 J. W. Hornsey, J. Soc. Chem. Ind., 30 (1911), 730. 

2 Probably degrees Fahrenheit. 

3 P. Klason, G. von Heidenstam, and E. Norlin, Arkiv Kemi, Mineral. 
Geol., 3 [10] (1908), 1-17; Z. angew. Chem., 22 (1909), 1205; 23 (1910), 1252; 
Ullmann's Enzykl. Techn. Chem., 6 (1919), 438. 

4 P. Klason, J. prakt. Chem., 90 (1914), 439. 



438 



CHEMISTRY OF CELLULOSE AND WOOD 



to bring it to a temperature of 275°, the total heat represented 
would be 4995.6 Calories. The same calculations on the distilla- 
tion products represent 4636.4 + 262.5 = 4898.9 Calories. The 
total heat set free would be 96.7 Calories. 

4995.6 - 4898.9 = 96.7 Calories. 

The heat of reaction, however, would be 

4895 - 4636.4 = 258.6 Calories. 

Similar data 1 for various woods and celluloses are given below: 



Thermochemistry of the Distillation of Wood and Cellulose 


Substance 


Heat 
of 

combus- 
tion, 

Calories 


Heat of 
combus- 
tion of 
prod- 
ucts, 
Calories 


Heat 
taken 
up by 
prod- 
ucts, 
Calories 


Heat 
liber- 
ated, 
Calories 


Heat 
liberated as 
per cent of 

heat of 
combustion 
of sub- 
stance, 
per cent 


Cotton 

Pine wood 


4,188 
5,070 
4,170 
4,910 
4,300 
4,910 
4,550 
4,790 
4,350 


3,840.9 
4,605.6 
3,854.6 
4,547.2 
3,896.8 
4,470.4 
4,135.5 
4,322.5 
3,902.5 


288.1 
241.1 
280.1 
233.0 
262.9 
243.6 
278.2 
245.3 
281.7 


152.3 
316.6 
158.6 
223.3 
233.6 
289.3 
239.8 
315.5 
279.1 


3.6 
6.3 


Pine cellulose 


3.8 


Spruce wood 


4.6 


Spruce cellulose 

Birch wood 


5.4 
5.9 


Birch cellulose 

Beech wood 


5.3 
6.6 


Beech cellulose 


6.4 



Calorific Value. — The calorific values of wood given by Klason 
in the above table may be supplemented by those of other investi- 
gators (p. 439). Most of the early work is unreliable, due to the 
use of the method based on the reduction of litharge to lead, or to 
failure to correct for hygroscopic water. 

Very complete information on the course of wood distillation, 
with respect to the formation of primary and secondary products, 
and the exothermic reaction, was obtained by Klason, 1 by first 

1 P. Klason et al, Z. angew. Chem., 23 (1910), 1256. 

2 P. Klason, J. prakt. Chem., 90 (1914), 445. 



THE DISTILLATION OF CELLULOSE AND WOOD 



439 



Calorific Values of European Woods 1 

Species Calories 

Oak (Quercus pedunculata) 4 , 620 

Ash {Fraxinus excelsior) 4,711 

Hornbeam (Carpinus betulus) 4, 728 

Beech (Fagus silvatica), 60 years old 4,766 

Beech (Fagus silvatica), 130 years old 4,785 

Beech (Fagus silvatica), 100 years old 4,770 

Birch (Betula alba) 4 , 771 

Scotch pine (Pinus silvestris) 5 , 035 

E. Gottlieb, J. prakt. Chem., 136 (1883), 392. 

Calorific Values of American Woods 

Species Calories* 



\ i 



> 2 



Chestnut 4 , 632 1 

Chestnut, leached 4 , 661 

Pine 4,920 

Oak 4,764 

Hickory 4 , 704 

Cherry 4,801 

Birch 4,710 

Poplar... 4,811 

* To obtain B.t.u. per pound, multiply the values given in calories in the above tables by 
1.796. 

i H. C. Sherman and C. G. Amend, School Mines Quart., 33 (1911), 30. 
2 S. W. Parr and C. N. Davidson, J. Ind. Eng. Chem., 14 (1922), 935. 

conducting the distillation in a cathode light vacuum and then 
gradually increasing the pressure to atmospheric. The yield 
and composition of the tar obtained with a high vacuum differed 
greatly from that at ordinary pressure. The exothermic reac- 
tion was found to be due to a secondary decomposition of the tar 
or pitch, and is not related to the primary decomposition of the 
wood. Under a high vacuum the pitch distills also, the decom- 
position of the wood taking place approximately as follows : 

3C 10 H 5 O + 19H 2 + 3C0 2 + 3CO 
Primary 16.8% 6.5% 4.1% 
charcoal 
20.8% 
2C 4 2H 6 o0 2 8 -> j + 2.5CH 3 .COOH + H.COOH + CH3OH 
Wood 7.5% 2.2% 1.6% 

+ C36H43O16 + C 5 H 8 0. 

Pitch Formaldehyde, oils, etc 
36% 4.5% 



440 



CHEMISTRY OF CELLULOSE AND WOOD 



By distillation under normal pressure over a period of 14 days, 
the pitch decomposes as indicated below : 

C 36 H430 16 -> C 3 oH 2 o0 3 + 9H 2 + 2C0 2 + C 4 H 9 . 
Pitch Secondary 8% 4.3% Oils 

36% charcoal 2.8% 

21.1% 

When the distillation was prolonged to 14 days at atmospheric 
pressure, the yield of tar was reduced to 1.8 per cent. Under 
ordinary commercial conditions the yield of tar is about 10 per 
cent. The variation in the yield of tar under various pressures 
and distillation periods greatly influences the thermochemistry of 
the reaction. In the distillation of birch wood 1 there was 
absorption of heat at the low pressure. 

Thermochemistry of Carbonization of Birch Wood for Various 
Times and Pressures, Final Temperature 400° 





5-milli- 
meter 
mercury 
vacuum 


Ordinary pressure 
for 




8 hours 


14 days 


Heat of reaction, Calories 


+ 123.1 

+ 2.6 

- 47.6 

- 1.0 


+258.6 

+ 5.3 
+ 96.7 

+ 2.0 


-4-340.9 


In per cent of heat of combustion of 
wood 


+ 7.1 


Heat liberated, Calories 


+ 169.2 


In per cent of heat of combustion of 
wood 


+ 3.5 







On theoretical grounds, the heat generated by the exothermic 
reaction should be nearly sufficient to accomplish the distillation 
without the aid of external heat. This end is approached most 
nearly by the Stafford 2 process. Hot, dry comminuted wood is 
fed continuously into the top of a vertical retort and charcoal is 
removed from the bottom. No external heat is applied to the 
retort after the process is under way. 

1 P. Klason, I.e., 442. 

2 O. F. Stafford, U. S. P. 1380262 (1921); Chem. and Met. Eng., 28 (1923), 
441. 



THE DISTILLATION OF CELLULOSE AND WOOD 441 

Experiments on vacuum distillation have also been made by 
Aschan. 1 Birch distilled at 6 to 8 millimeters gave 6.9 per cent 
of acetic acid, which is about the normal yield. Acetone appar- 
ently was not formed and its occurrence in the usual distillates 
must be due to decomposition of acetic acid. The yield of char- 
coal was 45 per cent in comparison with the normal 33 per cent. 
Similar distillations were made on pine and fir. 2 The charcoals 
from the three species were almost identical. The conifers gave 
more formic acid than acetic acid, while the reverse was true of 
the hardwoods. 

The distillation of resinous woods in vacuo greatly improves 
the yield of turpentine. 3 

The presence of Z-glucosan in the distillate obtained from the 
distillation of wood under reduced pressure might be expected 
from the patent of Pictet; 4 Sarasin, 5 however, states that it could 
not be obtained from wood. 

Yields of Products. — The table on page 442 shows the yields 
of products obtained by Klason 6 and his associates for European 
species. 

1 O. Aschan, Brennstoff-Chemie, 2 (1921), 273; 4 (1923), 164. 

2 O. Aschan, Brennstoff-Chemie, 4 (1923), 129, 145. 

3 M. Adams and C. Hilton, J. Ind. Eng. Chem., 6 (1915), 378-382. 

4 A. Pictet, U.S.P. 1437615 (1922). 

5 J. Sarasin, Arch. sci. phys. nal., [IV] 46 (1918), 32. 

6 P. Klason et al, Z. angew. Chem., 23 (1910), 1252. 



442 CHEMISTRY OF CELLULOSE AND WOOD 

Yields of Products by the Distillation of European Woods 



Species 



Pine, 
per cent 



Spruce, 
per cent 



Birch, 
per cent 



Beech, 
per cent 



Composition of raw material 

Water 

Ash 

Organic matter 

Distillation products: 

Charcoal 

Carbon dioxide 

Ethylene 

Carbon monoxide 

Methane 

Methyl alcohol 

Acetone 

Methyl acetate 

Acetic acid 

Organic matter in 
CH 3 .COONa 

Tar 

Water 

Loss 



7.52 


9.27 


8.79 


0.21 


0.22 


0.38 


92.27 


90.51 


90.83 


37.83 


37.81 


31.80 


10.13 


10.30 


9.96 


0.23 


0.20 


0.19 


3.74 


3.76 


3.32 


0.59 


0.62 


0.54 


0.88 


0.96 


1.60 


0.18 


0.20 


0.19 


0.01 


0.02 


0.02 


3.50 


3.19 


7.08 


8.03 


7.75 


8.15 


11.79 


8.08 


7.93 


22.27 


25.70 


27.81 


0.82 


1.41 


1.41 



7.11 
0.43 

92.46 

34.97 
10.90 
0.20 
4.22 
0.47 
2.07 
0.20 
0.03 
6.04 

5.89 

8.11 

26.58 

0.32 



An extensive series of investigations on American hardwoods 
has been made by Hawley and Palmer. 1 The yields may be 
accepted as comparable to those obtained on an industrial scale. 

The portion of the tree from which the wood is taken is of 
influence, though there is no regularity between species. In 
general, the yields of acetic acid and methyl alcohol are greater 
from heartwood than from sapwood, tupelo, however, being a 
striking exception. The high yield of acetic acid from beech 
slabs is due to the sapwood. Maple bark gives almost the same 
yield of methyl alcohol as the slabwood and heartwood. 

Senff 2 obtained the best yields of valuable products from 
sound trunk wood; limbs, decayed wood, and bark were inferior. 

1 L. F. Hawley and R. C. Palmer, U. S. Dept. Agr. Bull, 129 (1914), and 
508 (1917). 

2 M. Senff, Ber., 18 (1885), 60. 



THE DISTILLATION OF CELLULOSE AND WOOD 



443 



In the case of chestnut, Borghensani 1 found that the yields of 
both acetic acid and methyl alcohol increased in the order: 
trunk, root, and branch. 



Yields of Products from Heartwood of Various Species 2 



Species 



Locality 



Methyl 
alcohol, 
per cent 



Acetic 

acid, 

per cent 



Char- 
coal, 
per 
cent 



Tar, 
per 
cent 



Beech 

Beech 

Birch 

Birch 

Maple 

Maple 

Red gum 

Chestnut 

Hickory 

White oak 

White oak 

Tupelo 

White elm 

Slippery elm 

Silver maple 

Green, blue, and yel- f 

low ash \ 

Black ash 

Chestnut oak 

Tanbark oak 

Swamp oak 

Eucalyptus 



Indiana 

Pennsylvania 

Wisconsin 

Pennsylvania 

Wisconsin 

Pennsylvania 

Missouri 

New Jersey 

Indiana 

Indiana 

Arkansas 

Missouri 

Pennsylvania 

Wisconsin 

Wisconsin 

Tennessee 

Missouri 

Wisconsin 

Tennessee 

California 

Louisiana 

California 



1.95 


5.56 


2.23 


5.77 


1.45 


6.71 


1.62 


6.19 


1.94 


5.42 


1.94 


5.66 


1.76 


5.70 


0.90 


5.50 


2.08 


5.05 


1.34 


4.97 


1.33 


4.23 


1.56 


4.49 


2.12 


6.39 


2.03 


5.77 


1.69 


6.30 


1.91 


4.64 


1.79 


5.65 


1.22 


4.88 


1.72 


6.89 


1.50 


4.90 


1.33 


4.58 



43.1 
40.6 
39.0 
36.4 
39.8 
40.2 
36.8 
47.6 
37.7 
49.5 
42.0 
45.9 
40.0 
40.7 
41.2 
41.0 

38.2 
39.6 
37.6 
46.5 

48.2 



9.7 

9.1 

9.6 

12.6 

12.4 

12.5 

11.7 

4.7 

13.0 

6.3 

9.3 

11.4 

12.1 

9.6 

12.1 

11.3 

11.4 

10.2 

9.0 

7.3 

3.7 



1 G. Borghensani, Chem. Ztg., 34 (1910), 609. 
2 L. F. Hawley and R. C. Palmer, I.e. 



444 CHEMISTRY OF CELLULOSE AND WOOD 

Yields of Products from Slabwood of Various Species 1 



Species 



Locality 



Methyl 
alcohol, 
per cent 



Acetic 

acid, 

per cent 



Char- 
coal, 
per 
cent 



Tar, 
per 
cent 



Beech 

Beech 

Birch 

Birch 

Maple 

Maple 

Red gum 

Chestnut 

White oak 

White oak 

Tupelo 

White elm 

Slippery elm 

Silver maple 

Green, blue, and yel 

low ash 

Black ash 

Chestnut oak 

Tanbark oak 

Swamp oak 

Eucalyptus 



Indiana 

Pennsylvania 

Wisconsin 

Pennsylvania 

Wisconsin 

Pennsylvania 

Missouri 

New Jersey 

Indiana 

Arkansas 

Missouri 

Pennsylvania 

Wisconsin 

Wisconsin 

Tennessee 

Missouri 

Wisconsin 

Tennessee 

California 

Louisiana 

California 



79 
09 
55 
59 
91 
78 
73 



0.87 
1.33 
1.46 
1.86 
1.68 
1.79 
1.77 



43 
04 
30 
53 
31 



6.18 
6.21 
6.88 
6.10 
5.11 
5.44 
5.23 
5.26 
4.77 
4.35 
8.19 
6.61 
5.53 
5.31 

4.14 
5.16 
4.91 
6.01 
5.43 
5.31 



39.4 
42.3 
38.1 
38.6 
44.9 
37.6 
47.4 
52.9 
50.5 
46.2 
46.1 
39.7 
44.0 
44.6 

45.8 
40.5 
46.9 
41.9 

47.4 
44.2 



10.6 

10.9 

8.5 

10.3 

9.2 

12.3 

7.5 

3.7 

4.6 

8.7 

12.1 

11.1 

7.0 

8.0 

7.9 
9.1 
8.8 
10.0 
8.9 
8.8 



Yields of Products from Limbs and Bark of Various Species 1 


Species 


Locality 


Form 


Methyl 
alcohol 


Acetic 
acid 


Beech 


Indiana 

Wisconsin 

Indiana 

New Jersey 

Missouri 

Missouri 

California 

California 


Bark 

Bark 

Sapwood 

Limbs 

Limbs 

Limbs 

Limbs 

Limbs 


1.25 

1.88 
1.97 
0.96 
1.64 
2.02 
1.66 


2.98 


Maple 


3.15 


Beech 


6.67 


Chestnut 


6.42 


Tupelo 


5.64 


Green ash 


4.51 


Tanbark oak 




Black oak 


6.76 







X L. F. Hawley and R. C. Palmer, I.e. 



THE DISTILLATION OF CELLULOSE AND WOOD 



445 



The softwoods have been studied sparingly in comparison 
with the hardwoods. The following data were obtained by 
Adams and Hilton. 1 Each charge consisted of 25 kilograms of 
wood. The distillation was conducted under diminished pres- 
sure until the temperature of the retort reached 250°. 



Species 


Pyrolig- 
neous 
acid, 
liters 


Methyl, 

alcohol, 

per 

cent 


Acetic 
acid, 
per 
cent 


Tar, 
liters 


Charcoal 




Kilos 


Per cent 


Sugar pine 


9.8 
12.4 
11.1 
10.1 


0.52 
0.55 
0.48 
0.51 


2.6 
2.2 
1.8 
2.1 


1.26 
2.09 

1.78 
1.23 


3.64 
4.32 
5.03 

4.77 


14.56 


Western yellow pine. . 
Red fir 


17.28 
20.12 


Silver fir 


19.08 









Distillation with Superheated Steam. — Following Violette, 2 
various attempts have been made to use superheated steam for 
distilling wood. This heating medium is claimed 3 to give higher 
yields of methyl alcohol and acetic acid, the latter attaining 
about 8 per cent with birch. The initial action of the steam is 
without doubt hydrolytic, most of the acetic acid being formed 
by hydrolysis of the acetyl groups. According to Biittner and 
Wislicenus, 4 decomposition of wood with superheated steam 
begins at 240°. Less methyl alcohol was obtained from birch 
than by the usual methods, but the yield of acetic acid was 
much higher, namely, about 10 per cent. Both lignin and cellu- 
lose gave acetic acid, but no methyl alcohol was obtained from 
cellulose. 

Effect of Moisture. — The amount of water present in wood 
influences the character of the distillate, though no general 
conclusions can be drawn, since the results vary with the species. 
Palmer and Cloukey 5 distilled beech, yellow birch, and maple 
which had been seasoned for periods of 18 months, and 4 to 5 

1 M. Adams and C. Hilton, /. Ind. Eng. Chem., 6 (1914), 378; cf. H. K. 
Benson and M. Darrin, Ibid., 7 (1915), 916. 

2 H. Violette, I.e. 

3 P. Poore, E. P. 131006 (1919). 

4 G. BfiTTNER and H. Wislicenus, J. prakt. Chem., 10 (1918), 177-234. 

5 R, C. Palmer and H. Cloukey, /. Ind. Eng. Chem., 10 (1918), 262-264. 



446 CHEMISTRY OF CELLULOSE AND WOOD 

months. It is unfortunate that there was not greater variation 
in the moisture content of the wood, this ranging from 65 to 100 
per cent (based on the dry weight). Moisture had a favorable 
effect on the yield of acetic acid with controlled distillations. 
In the case of beech, the highest yields of methyl alcohol and 
acetic acid were obtained by controlling the distillation of the 
wetter wood. Moisture had little influence on the yield of 
acetic acid from yellow birch, though a controlled distillation 
of the drier wood gave the most alcohol. Maple resembled 
birch in its behavior. The largest yield of formic acid was 
obtained by the rapid distillation of wet wood, particularly beech. 

Order of Formation of Products. — The order in which the 
various distillation products are formed is of interest. Palmer 1 
made a series of distillations, interrupted at definite points, on 
yellow birch, and found that, aside from gas, formic acid was 
the first to appear, followed by acetic acid, tar soluble in pyro- 
ligneous acid, methyl alcohol, and "settled tar." Birch, 2 con- 
taining about 10 per cent of water, gave 8 to 10 per cent of its 
total acetic acid and 1 per cent of its methyl alcohol before the 
exothermic reaction set in. According to Blacher, 3 when the 
distillation is conducted at a temperature not exceeding 260°, 
about 33 per cent of the normal amount of acetic acid is obtained, 
and 80 per cent up to a temperature of 300°. 

Klason 4 found that the yields of methyl alcohol and acetic acid 
remained practically constant, regardless of the speed of distilla- 
tion, under pressures varying from a high vacuum to atmospheric 
pressure. No acetone was produced in vacuo. It is formed at 
the expense of the acetic acid during a prolonged distillation. 
The highest yields of formic acid are obtained by vacuum distil- 
lation and the lowest by prolonged distillation at normal pres- 
sure. The same is true of formaldehyde except that the decrease 
is less. Klason obtained 1.27 per cent of formaldehyde by 
vacuum distillation and 0.80 per cent by a distillation at 
atmospheric pressure lasting 14 days. 

1 R. C. Palmer, J. Ind. Eng. Chem., 10 (1918), 260-262. 

2 L. F. Hawley and R. C. Palmer, U. S. Dept. Agr. Bull, 129 (1914), 5. 

3 C. Blacher, Chem. Ind., 23 (1910), 508. 

4 P. Klason, J. prakt. Chem., 90 (1914), 413. 



THE DISTILLATION OF CELLULOSE AND WOOD 447 



The distribution of acetic acid during the slow distillation of 
the trunk wood of pine was determined by Bornstein. 1 A charge 
consisted of 1 kilogram of wood containing 3.21 per cent of 
moisture. 





Pyroligneous 


acid, 


Acetic acid, 




grams 




grams 


To 250° 


65.0 




0.91 


250-300 


65.3 




3.32 


300-350 


145.7 




15.15 


350-400 


71.2 




5.48 


400-450 


20.9 




0.48 


Total 


25 . 34 = 2.5 per cent 





Non-combustible gases were obtained at 165°, water at 180°, 
combustible gases at 280°, the first tar at 300°, and sulphur 
compounds at about 280°. 

Pressure above Atmospheric. — Palmer 2 has studied the effect 
of distilling wood at high pressures. He used pressures mainly 
up to 150 pounds per square inch and a final temperature of about 
335°. At a pressure of 450 pounds, the exothermic reaction was 
very violent. Pressure produced a slight increase of methyl 
alcohol, an increase of charcoal and gas, and a decrease of acetic 
acid and tar. The effect of pressures up to 60 pounds was 
much greater than for higher pressures, particularly on the tar. 
At 60 pounds' pressure, the yield of tar was decreased 60 to 65 
per cent. 

Temperature Control. — We have seen that wood is not decom- 
posed appreciably below 250° and that at 275° an exothermic 
reaction takes place that may become violent. To avoid loss 
of methyl alcohol and acetic acid by secondary reactions, it is 
very essential that the main decomposition be controlled by 
heating slowly or reducing the external heat at the critical point. 

Prior to the discovery of the exothermic reaction, Senff 3 made 
some small-scale distillations by slow and rapid heating. Slow 
distillation gave greater yields of acetic acid than rapid. 

1 E. Bornstein, J. GasbeL, 49 (1906), 648. 

2 R. C. Palmer, J. Ind. Eng. Chem., 6 (1914), 890. 

3 M. Senff, Ber., 18 (1885), 60-64. 



448 



CHEMISTRY OF CELLULOSE AND WOOD 



Species 



Total 




distil- 


Tar, 


late, 


per 


per 


cent 


cent 




50.53 


6.39 


47.76 


7.06 


51.05 


5.46 


42.98 


3.24 


48.15 


3.70 


45.24 


3.20 


45.37 


4.42 


51.75 


9.77 


51.61 


9.30 


43.77 


5.58 



Acetic 


Char- 


acid, 


coal, 


per 


per 


cent 


cent 



Gas, 
per 
cent 



Alder (Alnus glutinosa L.) J a 

1 ^ 
Birch (Betula alba L.) J a 

)b 

Oak (Quercus robur L.) \ a 

)b 

Fir (Pinus abies L.) J a 

1 ^ 
Larch (Pinus larix L.) J a 

1 b 



5.77 


31.56 


4.13 


21.11 


5.63 


29.24 


4.43 


21.46 


4.08 


34.68 


3.44 


27.73 


2.73 


30.27 


2.39 


24.18 


2.69 


26.74 


2.06 


24.06 



17.91 
31.13 
19.71 
35.56 
17.17 
27.03 
24.36 
24.07 
21.65 
32.17 



o. Slow distillation, b. Rapid distillation. 

Somewhat similar experiments were made under Rudnew. 1 
Sawdust (20 to 47 grams) dried at 120° was distilled in a glass 
retort heated by a metal bath. After reaching 150°, the tempera- 
ture of the bath was raised 1° per minute to 300°; then the dis- 
tillation was completed without temperature measurements. It 
was concluded that most of the acetic acid came from the lignin 
and that, since linden gave the highest yield, the actual hardness 
of the wood was not the important factor. The yields of acid 
are suspiciously high. 





Linden 


Birch 


Aspen 


Oak 


Pine 


Fir 


Birch 
cellu- 
lose 


Pine 
cellu- 
lose 


Acetic acid, per cent. 


10.24 


9.52 


8.06 


7.92 


5.65 


5.24 


6.21 


5.07 



Barillot, 2 as a result of distillations made on a laboratory (100 
kilograms) and on a commercial scale, found that it was neces- 
sary to distil slowly and regularly in order to obtain maximum 
yields of acetic acid, methyl alcohol, and charcoal. Similar 
conclusions were reached by Borghensani, 3 whose data are 
given below: 

1 W. Rudnew, Dingier s poly tech. J., 264 (1887), 88-92. 

2 E. Barillot, Compt. rend., 122 (1896), 467-471, 735-736. 

3 E. Borghensani, Chem. Ztg., 34 (1910), 609-610. 



THE DISTILLATION OF CELLULOSE AND WOOD 449 
Effect of Rate of Distillation on Yields of Products from Chestnut 



Wood 


Rate 

of 
distil- 
lation 


Char- 
coal, 
per 
cent 


Tar, 
per 
cent 


Acetic 

acid, 

per 

cent 


Methyl 

alcohol, 

per 

cent 


Ace- 
tone, 
per 
cent 


Gas, 


Bark, 
per 

cent 


Water, 
per 

cent 


Form 


per 
cent 


5.45 
20.00 
20.10 


33.17 
17.37 
31.14 


Trunk 

Branch 

Root 


j Slow 
) Rapid 
j Slow 
) Rapid 
j Slow 
\ Rapid 


30.09 
25.74 
22.84 
19.54 
26.25 
22.74 


3.96 
3.30 
4.32 
3,60 
4.06 
3.38 


2.70 
2.25 
4.65 
3.90 
3.30 
2.75 


0.68 
0.56 
1.16 
0.97 
0.83 
0.69 


0.07 
0.06 
0.12 
0.11 
0.08 
0.07 


13.20 
23.19 
19.00 
27.96 
17.50 
26.59 



The rate of distillation was measured by the weight of distillate 
collected in a given time (0.5 kilogram per hour by slow distilla- 
tion), rapid distillation being 20 times faster than slow distillation. 
In this case the rate had very slight influence on the formation of 
acetone. 

The beneficial effects of temperature control have been 
strikingly brought out by Palmer. 1 Some of the results are 
given in the following table : 



Influence of Temperature Control on the Yields of Methyl Alcohol 

and Acetic Acid 




Methyl alcohol 


Acetic acid 


Species 


Uncontrolled, 
per cent 


Controlled, 
per cent 


Uncontrolled, 
per cent 


Controlled, 
per cent 


Maple 


1.56 
1.59 
1.62 
1.59 
2.01 
2.03 
2.07 
2.04 
1.59 
1.67 
1.63 


2.16 
2.36 
2.41 
2.31 
2.15 

2.15 
1.73 
1.76 
1.75 


5.29 
5.74 
5.91 
5.65 
5.70 
5.78 
5.85 
5.77 
6.58 
6.50 
6.54 


6.30 


Maple 


5.45 


Maple 


5.52 


Mean 


5.76 


Beech 


6.28 


Beech 




Beech 




Mean 


6.28 


Birch 


6.96 


Birch 


6.78 


Mean 


6.87 







1 R. C. Palmer, J. Ind. Eng. Chem., 7 (1915), 663. 



450 



CHEMISTRY OF CELLULOSE AND WOOD 



In all experiments, a final temperature of 350° was reached. The 
heating bath was so regulated in the controlled runs that the 
temperature at the center of the retort did not exceed 275 to 300° 
during the critical period. The various species were not equally 
benefited by controlled distillations. The increased yield of 
methyl alcohol was 45 per cent for maple, 6 per cent for beech, 
and 7 per cent for birch, while the increase in acetic acid for 
these species was 2, 9, and 5 per cent respectively. 

The critical point for methyl alcohol is just before the formation 
of the tar, 1 but regardless of how the distillation is made, the yield 
of methyl alcohol falls far short of what would be expected from 
the methoxyl content of the wood. This is shown in the follow- 
ing table : 2 





Birch 


Maple 


Western 
larch 


Methyl alcohol, per cent 


1.53 
6.07 


1.94 

7.25 


0.65 


Methoxyl, per cent 


4.95 







The low yields of methyl alcohol are due mainly to the formation 
of phenol ethers and methane. 

Use of Catalyzers. — Various attempts have been made to 
increase the yields of methyl alcohol and acetic acid by having 
various added chemicals present during the distillation. The 
claim 3 for increased yields by distilling wood with sulphuric acid 
could not be confirmed by Palmer. 4 He found that no methyl 
alcohol was formed and that the yields of acetic and formic acids 
were greatly reduced. 

The presence of phosphoric acid was beneficial. Palmer 4 
used from 1.25 to 23.75 per cent of phosphoric acid, and pressures 
from to 110 pounds. In one experiment, beech distilled with 
2.45 per cent of phosphoric acid at 110 pounds' pressure gave 
13.85 per cent of acetic acid. With both beech and maple, there 

1 R. C. Palmer, J. Ind. Eng. Chem., 10 (1918), 260. 

2 A. W. Schorger, J. Ind. Eng. Chem., 9 (1917), 566. 

3 Orljavacer Chem. Fabrik, F. P. 357432 (1905); G. P. 185934 (1905). 

4 R. C. Palmer, J. Ind. Eng. Chem., 10 (1918), 264; U. S. P. 1259277 
(1918). 



THE DISTILLATION OF CELLULOSE AND WOOD 451 

was a pronounced increase in the yield of methyl alcohol, but, 
in general, the yield of acetic acid was not improved. When 
tar was distilled with wood, under pressure, in the presence of 
phosphoric acid, the yield of methyl alcohol was greatly increased, 
owing to fission of the methoxyl groups in the tar ; at atmospheric 
pressure, no increase was obtained. The most pronounced effect 
of phosphoric acid was to decrease the soluble tar and reduce the 
settled tar to a negligible quantity. 

On continuing this work, Hawley 1 found that phosphoric 
acid in concentrations of 0.5 to 1.5 per cent gave a slight decrease 
of methyl alcohol and a slight increase of acetic acid. 

Other chemicals used by Hawley 2 were lime, calcium carbonate, 
sodium carbonate, sodium silicate, magnesium oxide, ferric 
oxide, and magnesium chloride. The most promising results 
were obtained with low concentrations of sodium carbonate. 3 
The chemical must be uniformly distributed throughout the 
wood. Maple with 1.5 per cent and white oak with 0.5 per cent 
sodium carbonate gave a large increase of methyl alcohol without 
materially affecting the yield of acetic acid. The yields were: 





Methyl 
alcohol, 
per cent 


Acetic 

acid, 

per cent 


Silver maple: 

Untreated 


1.61 
2.39 

1.17 

2.58 


5.22 


Treated 


5.26 


White oak: 

Untreated , . 


4.91 


Treated 


5.09 







Distillation of cellulose and wood with 20 to 80 per cent of 
sodium hydroxide increased the yield of gas and tar. 4 Zinc 
chloride, apparently due to its dehydrating action on the wood, 
reduced the yield of tar and greatly increased the amount of 
charcoal. 



1 L. F. Hawley, J. Ind. Eng. Chem., 14 (1922), 43. 

2 L. F. Hawley, I.e. 

3 L. F. Hawley, U. S. P. 1363730 (1920). 

4 F. Fischer and H. Niggemann, Abhandl, Kennt. Kohle, 1 (1917), 176-183. 



452 



CHEMISTRY OF CELLULOSE AND WOOD 



Fremy, 1 in 1835, distilled sugar, starch, and gum with eight 
parts of lime and obtained a mixture of ketones consisting 
mainly of acetone. 2 Basset 3 states that wood slowly distilled 
with only 2.5 parts of lime gave 26 per cent of acetone and higher 
ketones. With a smaller quantity of lime (0.6 part), Hawley 4 
obtained a marked decrease in the yield of acetic acid without 
an increase of acetone. Basset's results could not be confirmed 
by Schorger, 5 who obtained about 2 per cent of acetone. 

When maple wood is distilled with sodium carbonate, the 
increased yield of methyl alcohol is at the expense of the methoxyl 
groups in the charcoal and dissolved tar. 6 The increase, in the 
case of oak, can be balanced only by taking into consideration 
the decrease in methane in the gas. It will be noted that only 
about 70 per cent of the original methoxyl is accounted for in 
the table below. 

Distribution of Methoxyl Groups in the Products of the Distilla- 
tion of Maple and Oak Wood 



Product 



Maple a 



Blank, 
per 
cent 



1 per 

cent 

Na 2 C0 3 , 

per cent 



White oak 6 



Blank, 
per 

cent 



1 per 

cent 

Na 2 C0 3 , 

per cent 



Pyroligneous acid 

Dissolved tar 

Settled tar 

Charcoal 

Gas (CH 4 calculated as CH 3 0) 

Total methoxyl 



1.62 
0.34 
0.52 
0.28 
1.31 

4.07 



1.92 
0.21 
0.59 
0.04 
1.47 

4.23 



1.16 
0.22 
0.46 
0.70 
1.34 



1.84 
0.16 
0.60 
0.13 

0.78 

3.51 



a. Original maple had 6.09 per cent CH3O. b. Original oak had 5.12 per cent CH3O. 



1 E. Fremy, Ann., 15 (1835) 277; Ann. chim. phys., [2] 59 (1835), 5. 

2 Contrary to several statements, Fremy does not mention having used 
wood; nor does he give any yields of acetone. 

3 H. P. Basset, Chem. Met. Eng., 20 (1919), 190. 

4 L. F. Hawley, I.e. 

5 A. W. Schorger, J. Ind. Eng. Chem., 17 (1925), 944. 

6 L. F. Hawley and S. S. Aiyar, J. Ind. Eng. Chem., 14 (1922), 1056. 



THE DISTILLATION OF CELLULOSE AND WOOD 453 

It would be of interest to determine the effect of the natural 
mineral constituents of the wood on the distillation products. 
It is possible that some of the variations usually attributed to 
a difference in the composition of the lignocelluloses, especially 
where the same species has grown on different soils, may be due 
to the effects of the ash. 

Charcoal. — The yield and the composition of charcoal depend 
upon the species of wood, and upon the rate and temperature of 
distillation. Violette, l using steam, obtained the following yields : 



Yields of Charcoal at Various Temperatures 



Temperature, degrees Centigrade. 
Yield, per cent 



160 



170 
94.55 



180 190 200 210 230 250 270 
18.59 81.99 77.10 73.14 55.37 49.67 37.14 



Temperature, degrees Centigrade. 
Yield, per cent 



280 
36.16 



300 
33.61 



320 
32.23 



340 
31.53 



350 432 

29.66 18.87 



1023 
18.75 



1100 1250 
18.40 17.94 



Rapid distillation at 432° gave only 8.96 per cent of charcoal in 
comparison with 18.87 per cent by slow distillation. 

The influence of temperature on the composition 2 of the char- 
coals is shown below. The ash may be obtained by subtracting 



Compositions 


r of Ci 


IARCOALS M 


ade at Various Temperatures 


Temper- 
ature, 


Composition 


, per cent 


Temper- 
ature, 

degrees 
Centi- 
grade 


Composition 


, per cent 


degrees 
Centi- 
grade 


C 


H 


O + N 


C 


H 


O + N 


150 
160 


47.5 
47.6 


6.1 
6.1 


46.3 
46.3 


260 
270 


67.9 
70.5 


5.0 
4.6 


26.5 
24.2 . 


170 
L80 
190 


47.8 
48.9 
50.6 


6.2 
5.8 
5.1 


46.0 
45.1 
44.1 


280 
290 
300 


72.6 

72.5 
73.2 


4.7 
5.0 
4.3 


22.1 
22.0 
22.0 


200 


51.8 


4.0 


44.0 


310 


73.6 


3.8 


21.8 


210 


53.4 


4.9 


41.5 


320 


73.6 


4.8 


21.1 


220 


54.6 


4.2 


41.4 


330 


73.6 


4.6 


21.3 


230 


57.1 


5.5 


37.0 


340 


75.2 


4.4 


20.0 


240 
250 


61.3 
65.6 


5.5 

4.8 


32.7 
29.0 


350 
432 


76.6 
81.6 


4.1 

2.0 


18.4 
15.2 



1 H. Violette, Ann. chim. phys., [3] 32 (1851), 311. 

2 H. Violette, Ann. chim. phys., [3] 32 (1851), 322. 



454 



CHEMISTRY OF CELLULOSE AND WOOD 



the sum of the constituents from 100. The residues from heating 
at temperatures below 270° are very incompletely decomposed. 
The following table given by Klason 1 shows the effect of higher 
temperatures on the yield and composition of the charcoal. 



Yield and Composition of 


Charcoal at 


Various Temperatures 


Temperature, 
degrees 


Composition, per 


cent 


Yield, 








per 


Centigrade 


C 


II 


O 


cent 


200 


52.3 


6.3 


41.4 


91.8 


250 


70.6 


5.2 


24.2 


65.2 


300 


73.2 


4.9 


21.9 


51.4 


400 


82.7 


3.8 


13.5 


37.8 


500 


89.6 


3.1 


6.7 


33.0 


600 


92.6 


2.6 


5.2 


31.0 


700 


93.7 


2.4 


4.8 


28.7 


800 


95.8 


1.0 


3.3 


26.7 


900 


96.1 


0.7 


3.2 


26.6 


1000 


96.6 


0.5 


2.9 


26.5 



Rinman 2 passed nitrogen over coniferous charcoal heated to 
800°. Analysis of the gas evolved indicated that the charcoal 
had the following composition: 





C 


C0 2 


CO ' CH 4 


H 


Per cent 


87.0 


5.1 


5.3 


1.9 


0.7 







Wood charcoal contains 0.2 to 0.3 per cent of methoxyl, 3 
as shown by the Zeisel method. It has been found, also, that 
methyl alcohol 4 is formed when charcoal is heated at high 
temperatures, especially above 730°. 

The specific gravity of the charcoal examined by Werther 5 
varied from 1.45 to 1.53. Pickles 6 obtained the following values: 



1 P. Klason, "Ullmann's Enzykl. Techn. Chem.," 6 (1919), 443. 
2 L. Rinman, Dinglers polytech. J., 246 (1882), 472. 

3 L. F. Hawley and S. S. Aiyar, /. Ind. Eng. Chem., 14 (1922), 1056. 
4 L. F. Hawley, J. Ind. Eng. Chem., 15 (1923), 697. 

5 G. Werther, J. prakt. Chem., 61 (1854), 21. 

6 A. Pickles, Chem. News, 121 (1920), 1. 



THE DISTILLATION OF CELLULOSE AND WOOD 455 





Source 


Density 




Apparent 


Real 


Almond 


0.48 
0.23 
0.21 
0.63 


1.70 


Birch chips 


1.44 


Birch dust 


1.47 


Coconut 


1.71 







Charcoal, 1 after treatment with chlorine at a red heat, 1 had a 
specific gravity of 1.57 to 1.88. If the wood is subjected to 
great pressure during distillation, the apparent density is greatly 
increased. 2 

The heat of wetting of charcoal varies from 4.06 to 4.16 
calories per gram. 3 

The specific heat of pressed, powdered beech-wood charcoal 
at various temperatures has been determined by Kunz. 3 Weber 3 



Temperature, 

degrees 

Centigrade 


Specific heat 


Temperature, 

degrees 

Centigrade 


Specific heat 


435 
561 
728 
925 


0.243 
0.290 
0.328 
0.358 


1059 
1197 
1297 


0.362 
0.378 
0.381 



found the specific heat at lower temperatures to be: 

Temperature, Specific 

Degrees Centigrade Heat 

and 23 . 5 . 1653 

and 99 . 22 . 1935 

and 223 . 6 . 2385 

The calorific value of charcoal is high in comparison with 
wood, as will be observed from Klason's 4 data. The composition 

1 W. Luzi, Ber., 25 (1892), 1383. 

2 L. F. Hawley, J. Ind. Eng. Chem., 13 (1921), 301; U. S. P. 1369428 (1921). 

3 L. Kunz, Ann. Physik, 14 (1904), 323; H. F. Weber, Pogg. Ann., 154, 
(1875), 560. 

4 P. Klason et at., Z. angew. Chem., 23 (1910), 1252, 1256. 



456 CHEMISTRY OF CELLULOSE AND WOOD 

Calorific Value of Wood and Charcoal 





Wood 

calorific 

value, calories 


Charcoal 


Species 


Composition, per cent 


Calorific 




C 


H 


o 


value, calories 


Pine 


5,070 
4,910 
4,910 
4,790 


82.5 
82.5 
82.2 
82.1 


4.0 
4.1 
3.8 
4.1 


13.5 
13.4 
14.0 
13.8 


7,685 


Spruce 


7,695 
7,680 
7,555 


Birch 

Beech 



of the charcoal agrees closest with the formula Ci 6 Hi O 2 . Owing 
to the avidity with which charcoal takes up moisture and oxygen, 
analyses must not be accepted as representing the charcoal as 
it left the retort. According to Nikitin, 1 the heat of combustion 
of 1 gram of charcoal may be closely calculated from the formula 

Heat of combustion = 80.51 C + 273.4 H, 

in which C and H represent the percentages of carbon and hydro- 
gen in the dry charcoal. When tested with Klason's data, the 
formula is not reliable. 

Charcoal that has been subjected to a current of 2200 amperes 
and 60 volts for 10 minutes retains its structural form but gives 
graphitic acid on oxidation. 2 Judging from conductivity 
measurements, 3 graphitization begins at about 1200°. 

Preparation 

Temperature, Conductivity, 

Degrees Centigrade Milliamperes 

700 0.004 

1150 67.00 

1200 210.00 

1400 340.00 

To obtain a graphite similar to the natural product, a temperature 
of about 3000° is usually employed. 4 

Wood charcoal is not affected by concentrated sulphuric acid 
in the cold. 5 Hot concentrated sulphuric and nitric acids pro- 

i N. I. Nikitin, C. A., 11 (1917), 746. 

2 H. Moissan, Ann. chim. phys., [7] 8 (1896), 323. 

3 H. Herbst, Biochem. Z., 115 (1921), 213. 

4 C. Arsem, Trans. Am. Electrochem. Soc, 20 (1911), 110. 

5 M. Berthelot, Ann. chim. phys., [7] 14 (1898), 206. 



THE DISTILLATION OF CELLULOSE AND WOOD 457 

duce benzene polycarboxylic acids by oxidation. Boiling con- 
centrated sodium hydroxide dissolves charcoal with the evolution 
of hydrogen but not of carbon monoxide. 1 

Freshly prepared charcoal is highly combustible. It appears 
to absorb oxygen for 36 hours after carbonization. If powdered 
and stored before the expiration of this time, it will take 
fire spontaneously when exposed to the air. 2 Jacobs 3 has 
prepared from soda pulp "black liquor," a form of carbon that is 
highly reactive towards oxygen. It unites with atmospheric 
oxygen when heated to 150°; with some samples, at as low as 80°. 
The continuous absorption of oxygen by charcoal at ordinary 
temperatures over long periods of time results essentially in the 
formation of a solid oxide of carbon. 4 Heating produces decom- 
position with the formation of carbon monoxide and carbon 
dioxide. 

It had long been known that wood treated with various 
chemicals, such as acids and acid salts, charred, and washed gave 
a charcoal with greatly increased adsorptive powers. 3 It is 
stated 6 that the activated charcoal used in German gas masks 
was prepared by treating wood with a small amount of hydro- 
chloric acid and zinc chloride, charring, and washing. 

Opinions differ as to the reasons for the high activity of a 
charcoal. The amount and nature of the ash greatly influences 
the development of surface activity. According to Herbst, 7 
high adsorptive power is a specific property of pure, free, chem- 
ically unsaturated carbon. Wood charcoal shows increased 
adsorption with increase in carbon content and decrease in 
hydrogen. The adsorptive power for gases increases on heating 
up to 1150° but decreases above 1200°. 

According to Chaney, 8 all "primary" amorphous carbons 
consist of a stabilized complex of hydrocarbons adsorbed on a 

1 F. Haber and L. Bruner, Z. Elektrochem., 10 (1894), 697. 

2 A. Hargreaves, J. Chem. Soc, 12 (1874), 420. 

3 C. B. Jacobs, Chem. Age, 31 (1923), 475; U. S. P. 1462752 (1923). 

4 H. H. Lowry and G. A. Hulett, J. Am. Chem. Soc, 42 (1920), 1393. 

5 H. L. Melsens, Compt. rend., 79 (1874), 375; R. Ostrejko, E. P. 
14224 (1900). 

6 F. H. Carr, J. Soc. Chem. Ind., 38 (1919), 468R. 

7 H. Herbst, Biochem. Z., 115 (1921), 205. 

8 N. K. Chaney, Trans. Am. Electrochem. Soc, 36 (1919), 91. 



458 



CHEMISTRY OF CELLULOSE AND WOOD 



base of active carbon. Small quantities of a hydrocarbon, b.p. 
360°, were isolated from cedar-wood charcoal that had been 
heated to 850°. To obtain a highly active carbon, it is necessary 
to remove the hydrocarbons; though the removal can be accom- 
plished by high temperatures, the carbon is rendered inactive. 
It is possible to break down the hydrocarbons below the point at 
which inactive carbon is formed by means of air, steam, carbon 
dioxide, salts, and other chemical reagents. 

The nature of the medium in which the charcoal has been 
heated greatly affects the activity, since a small amount of the 
gas is retained as a complex. 1 In the following table the activity 
was measured by the amount of phenol taken up from a 1 per cent 
solution at 25° by 100 grams of alder-wood charcoal. 





Medium 




N 


CO 


CI 


Steam 


C0 2 


NH 3 


S 


SO* 


Air + N 


Temperature, degrees Centi- 


1200 
1.5 


1000 
1.5 


'•1000 



850 
25 


850 
25 


900 
35 


1000 



675 
30 


300 




15 











It is necessary to treat the carbon in different ways in order to 
obtain maximum adsorption of any particular compound. 

Constitution and Composition of Charcoal. — It has been 
concluded from X-ray examinations that amorphous carbon is 
identical with graphite, differing from it only in its greater degree 
of subdivision. 2 There is at least one marked chemical differ- 
ence. On oxidation, graphite gives graphitic acid and charcoal 
mellitic acid, mainly. 

Mellitic acid 3 is a characteristic product of the oxidation of 
most wood charcoals. Verneuil, 4 by means of concentrated 
sulphuric acid, obtained mellitic, benzene pentacarboxylic, and 
benzene tetracarboxylic acids. On oxidation with fuming nitric 
acid, 5 charcoal gave 25 per cent of ammonium mellitate. No 

1 O. Ruff et at., Kolloid-Z., 32 (1923), 225. 

2 P. Debye and P. Scherrer, Physik. Z., 18 (1917), 291. 
3 F. Schulze, Ber.,4: (1871), 801. 

4 A. Verneuil, Compt. rend., 118 (1894), 195; 132 (1901), 1340; cf. 
E. Philippt and R. Thelen, Ann., 428 (1922), 296. 

5 M. A. Dickson and T. H. Easterfield, Proc. Chem. Soc, 14 (1898), 
163; cf. E. P. 24662 (1907); E. Philippi and G. Rie, Ann., 428 (1922), 287 



THE DISTILLATION OF CELLULOSE AND WOOD 459 

difference was noted in the yield, if the charcoal contained 2.0 or 
0.2 per cent of hydrogen. 

Since charcoal yields mellitic acid, it has been assumed that the 
carbon exists as six-membered rings, each carbon atom being 
attached to carbon atoms without the ring. The barium salts of 
the benzene carboxylic acids, obtained by oxidizing charcoal with 
nitric acid, gave on distillation benzene, napthalene, and fluorene. 
It was concluded from the presence of fluorene that charcoal also 
contains bodies with a five-membered ring. 1 Various structures 2 
have accordingly been assigned to the carbon molecule. Aschan 3 
considers charcoal to be a hydrocarbon of high molecular weight, 
having the formula (CisoHio)* or (CmHioO)*. He proposes a 
honeycomb ring structure and suggests that the dark color of 
charcoal may be due to quinoidal linkages. 

It is reasonably certain that ordinary charcoal contains, in 
addition to free carbon, quantities of hydrocarbons as a residue 
from incomplete distillation. If charcoal is subjected to a white 
heat, oxidizing agents attack it very slowly and the yield of 
mellitic acid is almost nothing. 4 Meyer 5 concludes that the 
formation of mellitic acid has no bearing on the constitution of the 
carbon molecule. The jdeld of mellitic acid from different wood 
charcoals varied from to 40 per cent, the highest yields being 
obtained from coniferous charcoal. 

Composition of Distillates. — A large number of compounds 
have been identified in the pyroligneous acid and tar. These 
compounds are listed on pages 465-467. Most of them occur 
in very small amounts. For example, crude methyl alcohol 
does not contain more than 0.2 per cent of allyl alcohol, 
as was determined by refining 50,000 pounds of the crude alco- 
hol. 6 The fraction high in allyl alcohol, obtained in the refining 
process, is one of the few cheap sources for this alcohol. 



1 O. Dimroth and B. Kerkovius, Ann., 399 (1913), 120. 

2 J. Dewar, Chem. News, 97 (1908), 16; H. S. Redgrove, Ibid., 37; 
J. C. Thomlinson, Ibid., 37; H. E. Armstrong, J. Soc. Chem. Ind., 24 
(1905), 478. 

3 O. Aschan, Chem. Ztg., 33 (1909), 561-562. 

4 A. Verneuil, Compt. rend., 132 (1901), 1340. 

5 H. Meyer, Monatsh., 35 (1914), 163. 

6 M. Grodzki and G. Kraemer, Ber., 7 (1874), 1494. 



460 CHEMISTRY OF CELLULOSE AND WOOD 

The composition of pyroligneous acid and wood tar was long 
hedged about with uncertainty. 1 Many of the supposedly 
definite compounds were mixtures; thus, dumasin is a mixture 
of methyl-a-butenylketone and cyclopentanone. 2 Hermilian 3 
reported the presence of ethyl alcohol in crude methyl alcohol, 
but it could not be detected in the fraction, b.p. 70 to 100°, 
from 24,000 kilograms of raw material. 4 Though frequently 
reported, Fraps 5 could not find mesityl oxide or other condensa- 
tion products of acetone in the oil from hardwood tar. He reports 
minute quantities of nitriles. 

The oils from hardwood tars contain small quantities of sulphur 
compounds, possibly thiophenols. 6 Pastrovitch 7 isolated a 
phenol (coerulignol), Ci Hi 4 O 2 , that is possibly propyl-pyrogallol. 

Pyroxanthine (difurfural-ketopentamethylene) does not exist 
in the original pyroligneous acid. It is produced by condensa- 

C4H3O.CH : C— CH 2 .CH 2 — C : CH.OC4H3. 

' CO 1 

tion 8 of furfural with adipic ketone when the pyroligneous acid 
is treated with lime. 

The wine-red color produced on neutralizing pyroligneous acid 
with lime is obtained only when the lime contains traces of 
ferric salts. The indicator probably consists of the ethers of 
pyrogallol and its homologs. 9 

Oxidation of the higher phenol ethers gives dimethoxy- 
quinone and coerulignone. 10 Coerulignone (cedriret) has the 
composition Ci 6 Hi 6 6 and is apparently a tetramethoxy-p, p'- 

1 An interesting history of wood creosote is given by C. S. Schorlemmer, 
/. Soc. Chem. Ind., 4 (1885), 152-157. 

2 H. Pringsheim and J. Leibowitz, Ber., 56 (1923), 2034. 

3 V. Hermilian, Ber., 8 (1875), 661, 768. 

4 Verein f. Chemische Ind. zu Frankfort, Chem. Ztg., 20 (1896), 1015. 
s G. S. Fraps, Am. Chem. J., 25 (1901), 28. 

6 A. Behal and E. Choay, Com-pL rend., 118 (1894), 1339. 

7 P. Pastrovitch, Monatsh., 4 (1883), 188. 

8 D. Vorlander and K. Hobohm, Ber., 29 (1896), 1838. 

9 J. M. Johlin, J. Ind. Eng. Chem., 7 (1915), 596. 

10 C. Liebermann, Ber., 5 (1872), 746; 6 (1873), 381; 8 (1875), 69; Ann., 
169 (1873), 221; A. W. Hofmann, Ber., 7 (1874), 78. 



THE DISTILLATION OF CELLULOSE AND WOOD 461 

diphenjd quinone, though there are several possible structures. 1 
Dimethoxy-quinone has been prepared by the oxidation of 
propyl-pyrogallol-dimethyl-ether 2 and pyrogallol-trimethyl- 
ether. 3 

Hardwood tars are usually richer in phenols, especially guaia- 
col, than the softwood tars ; however, the tar from the distillation 
of redwood 4 is reported to possess an usually high phenol content. 

Medicinal wood creosote is usually prepared from beech, 
birch, and maple. It should distil mainly between 200 and 210°, 
though the composition will vary with the species. Behal and 
Choay 5 found that beech- wood creosote contained more guaiacol 
and less monophenols than oak-wood creosote. The phenol 
fraction of a beech- wood creosote, b.p. 200 to 210°, had the 
following composition: 

Per Cent Per Cent 

Phenol 5.20 1, 3, 4-Xylenol " 2.00 

o-Cresol 10.40 1, 3, 5-Xylenol 1.00 

m- and p-Cresols 1 1 . 60 Various phenols 6.20 

o-Ethylphenol 3 .06 Creosol and homologs 35.00 

Guaiacol 25 . 00 

The phenol fraction, b.p. 190 to 230°, from a pine tar examined 
by Renard 6 contained: 

Per Cent 

Monophenols 40 . 

Guaiacol 20 . 3 

Creosol and homologs 37 . 5 

Wood creosote is prepared by extracting the phenols from the 
heavy tar oils, b.p. 190 to 240°, with alkali. The best quality 
of creosote is obtained by extracting with an amount of 5 per 
cent sodium hydroxide such that only about two-fifths of the 
phenols present are dissolved. 7 

1 C. Liebermann and J. Flatau, Ber., 30 (1897), 234; A. W. Hofmann, 
Ibid., 11 (1878), 801. 

2 A. W. Hofmann, Ber., 11 (1878), 329. 

3 W. Will, Ber., 21 (1888), 608. 

4 W. H. Hund, U. S. P. 1365407 (1921). 

5 A. Behal and E. Choay, Compt. rend., 119 (1894), 166; cf. L. F. Kebler, 
Am. J. Pharm., 71 (1899), 409-413. 

6 A. Renard, Compt. rend., 119 (1894), 1276. 
7 L. F. Hawley, U. S. P. 1199271 (1916). 



462 



CHEMISTRY OF CELLULOSE AND WOOD 



The antiseptic properties of wood smoke are attributed to 
aliphatic aldehydes, especially formaldehyde. The smoke from 
100 grams of wood contained 0.5 gram of formaldehyde. 1 

Spruce tar decomposed at a red heat gave benzene, toluene, 
napthalene, and anthracene. 2 

Wood tar contains small amounts of the higher fatty acids. 
Bornstein 3 obtained 4 grams of palmitic acid and 2 grams of 
oleic acid from 1 kilogram of pine tar. According to Marcusson 
and Picard, 4 wood tar and pitch, particularly from hardwoods, 
are high in oxyacids. The pitch may also contain condensation 
products 5 of aldehydes, ketones, and phenols. 



Composition of Wood Tar and 


Pitch 








Unsapon- 
ifiable, 
per cent 


Anhy- 
drides of 
oxyacids, 
per cent 


Oxyacids 


Resin 

acids, 

per 

cent 


Fatty 

acids, 

per 

cent 




Product 


Ether 
insol- 
uble, 
per 
cent 


Ether 
sol- 
uble, 
per 
cent 


Phenols, 
per 
cent 


Beech tar 

Beech pitch 

Pine tar 


18 
6 
53.5 
19.7 


9.5 

7 



33.3 

7 

1 1 


19 
4 


7.7 


17 
35.2 


3.2 
1.5 
6 

2.8 


9.3 
1.5 
9.5 


Pine pitch 


31 


.8 




8.0 



The distillate from resinous woods contains resin acids and 
terpenes, and their decomposition products. Resin acids and 
terpenes are secondary constituents of wood. The terpenes will 
vary with the particular species of wood. Adams 6 found the oils 
from the wood of Pinus monophylla, Pinus jeffreyi, and Pinus 
ponderosa to contain practically the same constituents as Schorger 
had identified in the oleoresins of the same species; however, 
the volatile oils from the wood and oleoresin are seldom 
identical. The oil from the oleoresin usually consists entirely 



1 G. Pasqualis, Chem. Zentr., 68, II (1897), 1012. 

2 A. Atterberg, Ber., 11 (1878), 1222. 

3 E. Bornstein, J. GasbeL, 63 (1920), 90; cf. G. P. 314358 (1917). 

4 J. Marcusson and M. Picard, Z. angew. Chem., 34 (1921), 201-204; 
Chem. Umschau Fette, Oele, Wachse u. Harze, 28 (1921), 257-258. 

5 R. P. Duchemin, Bull Soc. Chim., 7 (1910), 473. 

6 M. Adams, J. Ind. Eng. Chem., 7 (1914), 957-960. 



THE DISTILLATION OF CELLULOSE AND WOOD 463 

of hydrocarbons. In the course of time the ter penes in the heart- 
wood undergo rearrangement and hydration with the formation of 
volatile oils of entirely different properties, such as the "wood 
turpentine" and "pine oil" distilled from the wood of the 
southern pines. Teeple 1 found that pine oil consisted mainly 
of a-terpineol. A thorough investigation of this oil by Schimmel 
and Company 2 revealed the following constituents: 

a-Pinene Camphor 

/3-Pinene Methyl-chavicol 

Camphene Borneol 

Z-Limonene a-Terpineol 

Dipentene Fenchyl alcohol 

7-Terpinene Isofenchyl alcohol 
Cineol 

A list of the numerous decomposition products of rosin, and the 
terpenes, obtained on distilling resinous wood from various 
species, does not fall within the province of this work. 

Source of Products. — It is useless to hazard an opinion on the 
groupings within the cellulose and lignin molecules that give rise 
to the products of wood distillation. The former confidence in 
pyrolytic decomposition as a method for determining constitu- 
tion is now deservedly wanting; however, Klason 3 has advanced 
a formula for lignin by which he explains the formation of phenols 
and phenol ethers, 2-methyl-furan, and allyl and methyl alcohol. 

The source of some products can be definitely assigned to 
cellulose or lignin by the process of elimination. Cellulose 
gives only small amounts of phenols, so that the large number of 
phenols and phenol ethers obtained from wood have their origin 
in the lignin. Lignin is profoundly modified by all available 
methods for its isolation, so that distillation of the modified 
product does not give reliable information as to what is obtain- 
able from the original. 

Methyl alcohol is derived solely from methoxyl groups in the 
lignin, cellulose yielding no methyl alcohol on distillation. 
Allyl alcohol, also, is derived from the lignin. 

1 J. E. Teeple, J. Am. Chem. Soc, 30 (1908), 412. 

2 Schimmel & Co. Report for April (1910), 104-109. 

3 P. Klason, Ber., 66 (1923), 300. 



464 CHEMISTRY OF CELLULOSE AND WOOD 

Acetic acid is formed from acetyl groups in the lignin and by 
decomposition of the cellulose. The presence of f ormyl groups in 
lignin has not been satisfactorily established. Formic acid is 
probably formed mainly by decomposition of the carbohydrates, 
hydrolysis first taking place in the presence of organic acids, at the 
prevailing high temperature. The hexose sugars would decom- 
pose partially into formic and laevulinic acids. 

C 6 H 12 6 -> CH 3 .CO.CH 2 .COOH + H.COOH + H 2 0. 
Laevulinic acid Formic acid 

Laevulinic acid on distillation gives the lactone of angelic acid, 
and on reduction, valeric acid. 

Aschan 1 has suggested that some of the aliphatic acids are 
formed from the hexoses as follows : 

C 6 H 12 6 -> 3CH 3 .COOH. 
Acetic acid 

C 6 H 12 6 -+ CH 3 .CH 2 .COOH + CH3.COOH + H.COOH. 
Propionic acid Acetic acid Formic acid 

C 6 H 12 6 -> CH 3 .CH 2 .CH 2 .COOH + 2H.COOH. 
Butyric acid 

Glucose on destructive distillation 2 gives formic and acetic 
acids, formaldehyde, acetone, methylfuran, furan, furfural, 
carbon dioxide, carbon monoxide, and methane. 

Furfural is formed mainly from pentosans. Furan and its 
homologs occur in the distillate from cellulose. It is not known 
if lignin free from carbohydrates will give furan, though a furan 
structure has been suggested for lignin. 

The ketones are probably formed by a secondary decomposi- 
tion of the fatty acids. The small amount of protein in wood 
furnishes the ammonia and pyridine. 

Wood on distillation yields larger quantities of methane than 
does cellulose. This must be due to the formation of methane 
from methoxyl groups. 3 

1 0. Aschan, Z. angew. Chem., 26 (1913), 711. 

8 E. von Lippmann, "Die Chemie der Zuckerarten," I (1904), 301. 

•L. F. Hawley and S. S. Aiyar, J. Ind. Eng. Chem., 14 (1922), 1056. 



THE DISTILLATION OF CELLULOSE AND WOOD 465 

Constituents of Hardwood Pyroligneous Acid 1 

Acids and Lactones 

Formic 1, 6, 9 Caproic 9 

Acetic 1 Angelic 6 

Propionic 5, 6, 9 a- and /3-Crotonic 6 

n-Butyric . . 5, 6, 9 ( Dihydroxyoctoic 32 

n- Valeric 5,6,9 1 C 7 H 13 (OH) 2 .COOH 

Valerolactone 8 

Alcohols 

Methyl. 1, 2 f Methyl-allyl 33 

Allyl 3, 4 I CH 3 .CH(OH).CH: CH 2 

Esters 
Methyl acetate 2, 10 Methyl formate 10. 

Aldehydes and Derivatives 

Formaldehyde 7, 48 Isovaleraldehyde 32 

Acetaldehyde 4, 7, 10 Furfural 1, 11 

Dimethyl-acetal 2 2-Methylfurfural 12 

Trimethyl-acetaldehyde .... 32 Pyroxanthine 53, 54, 55 

Ketones 

Acetone 1, la, 2, 10 Adipic ketone (cyclopen- 

Diethyl ketone 33 tanone) 23, 24, 32 

Diacetyl "40 Methyl-a-butenyl ketone . 32 

Methylpropyl ketone ... 22 Methyl-cyclopentenolone 28 

Methylethyl ketone 22, 31, 32 f 2-Keto-3-hexene 32 

Succinyl ketone 23 I CH 3 .CO.CH: CH.CH 2 .CH 3 

Bases 
Ammonia 1 Pyridine 23 

Hydrocarbons 

Toluene 19, 22 Cumene 19 

Xylene 19, 22 

Furans 

a, a-Dimethyl-tetrahydrof uran 33 

a-Methyl-a-ethyl-a, /3-dihydrofuran 33 

1 Figures are to references at end of chapter. 



466 CHEMISTRY OF CELLULOSE AND WOOD 

Constituents of Hardwood Tar 
Acids and Lactones 

Dokozanic 36 



Lignoceri0 29 ' 52 l C 21 H 4 ,0 2 or C 22 H„0 2 

Alcohols 
Isoamyl 20 Isobutyl 21 

Esters 

Methyl acetate 13 Methyl-n-butyrate 13 

Methyl propionate 13 Methyl-n-valerate 13 

Aldehydes 
Valeraldehyde 13 Propionaldehyde 16 

Ketones 

Methyl-cyclohexenone 26 

HC CH 2 CO 

Methylethyl ketone 13 ' 

Methyl-n-propyl ketone 13 y H q C(CH 3 ) CH 2 

Dimethyl-cyclohexenone 27 

HC -CH 2 CO 



H 3 C.C CH 2 CH.CH S 



Methyl-n-butyl ketone 13 

Diethyl ketone (?) 13 

Adipic ketone 13 * 

Pimelic ketone (cyclohexa- 

none) 21 

Methyl-cyclopentenone 20, 25 

H 2 C CO C.CH 3 

I II 

H 2 C CH 

Bases 
Pyridine 20 Dimethyl-pyridine 20, 21 

Furans 

a-Methyl-furan 13, 14 Trimethyl-furan 13 

Dimethyl-furan 13, 16 1, 2-Acetyl-furan 25 

Phenols 

Phenol 17, 18, 30, 37 Catechol 49 

o-Cresol 37 Guaiacol 17, 18, 30, 38 

ra-Cresol 37 1,3, 4-Ethyl-guaiacol . 37 

p-Cresol 17,18,30,37 Methyl-catechol- 

dimethyl-ether 30a 

o-Ethyl-phenol 37 Pyrogallol-dimethyl- 

ether 30,41,44 

1, 3, 4-Xylenol 17, 30, 37, 39 Methyl-pyrogallol- 

dimethyl-ether 30, 41, 45 

1, 3, 5-Xylenol 37 Prop y 1-pyrogallol- 

dimethyl-ether 30, 41, 44, 46 

Creosol 17, 30, 38 



THE DISTILLATION OF CELLULOSE AND WOOD 467 

Constituents of Softwood Pyroligneous Acid and Tar 

Acids 

Formic 50 Caprylic 50 

Acetic 50 Oleic 47 

Propionic 50 Palmitic 47 

n-Butyric 50, 51 Arachidic 47 

Valeric 42, 50 Abietic 47 

a- Methyl-valeric 50 Pimaric 42 

Caproic 42, 50 

(Enanthylic 42, 50 

Alcohols 
Methyl 51 AIM 51 

Esters 
Methyl-isobutyrate 34 

Aldehydes 

Acetaldehyde 35 Furfural 34 

Propionaldehyde 35 

Ketones 
Acetone 35, 51 Diacetyl 34 

Furans 

Furan 34 Dimethyl-f uran 34 

a-Methyl-furan 15, 34 

Hydrocarbons 

Benzene 34 m- Xylene 34 

Toluene 34 Retene 43, 50, 51 

Phenols 

Phenol 50 Guaiacol 42, 50 

Cresol 50 1, 3, 4-n-Propyl-guaiacol 42, 50 

Creosol 42, 50 1, 3, 4-Ethyl-guaiacol 42, 50 

References 

1. C. Volckel, Ann., 86 (1853), 66-113. 
la. C. Volckel, Ann., 80 (1851), 310. 

2. W. Dancer, Ann., 132 (1864), 240-243. 

3. B. Aronheim, Ber., 7 (1874), 1381-1382. 

4: M. Grodzki and G. Kraemer, Ber., 7 (1874), 1492-1497. 

5. T. Anderson, Chem. News, 14 (1866), 257. 

6. G. Kraemer and M. Grodzki, Ber., 11 (1878), 1356-1362. 



468 CHEMISTRY OF CELLULOSE AND WOOD 

7. R. P. Duchemin, Bull, soc. chim., [4] 7 (1910), 477. 

8. M. Grodzki, Ber., 17 (1884), 1369. 

9. Barre, Compt. rend., 68 (1869), 1222-1224. 

10. C. F. Mabery, Am. Chem. J., 5 (1883), 256-263. 

11. H. B. Hill, Ber., 10 (1877), 936-993. 

12. H. B. Hill, Ber., 22 (1889), 607; H. B. Hill and W. L. Jennings, 
Am. Chem. J., 15 (1893), 159. 

13. G. S. Fraps, Am. Chem. J., 25 (1901), 26-53. 

14. C. Harries, Ber., 31 (1898), 37-47. 

15. A. Atterberg, Ber., 13 (1880), 879-883. 

16. E. Fischer and W. J. Laycock, Ber., 22 (1889), 105. 

17. S. Marasse, Ann., 152 (1869), 59-87; Ber., 1 (1868), 99-100; Ibid., 
2 (1869), 71-73. 

18. E. von Gorup-Besanez, Ann., 86 (1853), 223-258; Ibid., 143 (1867), 
129-174. 

19. A. Cahours, Ann., 76 (1850), 286-287. 

20. E. Looft, Ann., 275 (1893), 366-382. 

21. E. Looft, Ber., 27 (1894), 1542-1546. 

22. Vladesco, Bull. soc. chim., 3 (1890), 510-514. 

23. W. Hentzschel, Ann., 275 (1893), 318-322. 

24. D. VoRLANDERand K. Hobohm, Ber., 29 (1896), 1836; H. METZNERand 
D. Vorlander, Ibid., 31 (1898), 1885-1886. 

25. L. Bouveault, Compt. rend., 125 (1897), 1184-1186. 

26. A. Behal, Compt. rend., 126 (1898), 46-49; cf. Ibid., 125 (1897), 1036- 
1038. 

27. A. Behal, Compt. rend., 132 (1901), 342-345. 

28. J. Meyerfeld, Chem. Ztg., 36 (1912), 549-552. 

29. C. Hell, Ber., 13 (1880), 1709-1713; Ibid., 1713-1751. 

30. F. Tiemann and P. Koppe, Ber., 14 (1881), 2005-2014. 

30a. F. Tiemann and B. Mendelsohn, Ber., 8 (1875), 1136-1139. 

31. G. Kraemer and M. Grodzki, Ber., 9 (1876), 1920-1927. 

32. H. Pringsheim and J. Leibowitz, Ber., 56 (1923), 2034-2041. 

33. H. Pringsheim and A. Gorgas, Ber., 57 (1924), 1561-1566. 

34. O. Aschan, Z. angew. Chem., 20 (1907), 1811-1816. 

35. O. Aschan, Z. angew. Chem., 26 (1913), 709-713. 

36. I. K. Traubenberg, J. Soc. Chem. Ind., 42 (1923), 303A. 

37. A. Behal and E. Choay, Compt. rend., 118 (1894), 1339-1342; 119 
(1894), 166-169. 

38. H. Hlasiwetz, Ann., 106 (1858), 339. 

39. M. Pfranger, Arch. Pharm., 228 (1890), 713-719. 

40. J. M. Johlin, J. Am. Chem. Soc, 37 (1915), 892. 

41. E. J. Pieper, S. F. Acree, and C. J. Humphrey, J. Ind. Eng. Chem., 
9 (1917), 462-465. 

42. M. Nencki and N. Sieber, Arch. Exp. Path. Pharm., 33 (1894), 17. 

43. A. Wahlforss, Chem. Zentr., 40 (1869), 479. 

44. A. W. Hofmann, Ber., 11 (1878), 329-338; cf. Ibid., 8 (1875), 66. 

45. A. W. Hofmann, Ber., 12 (1879), 1371. 



THE DISTILLATION OF CELLULOSE AND WOOD 469 

46. G. Niederist, Monatsh., 4 (1883), 487-493; cf. P. Pastrovitch, Ibid., 
182-187. 

47. E. Bornstein, /. Gasbel., 63 (1920), 90. 

48. P. Klason, J. prakt. Chem., 90 (1924), 413. 

49. M. Pettenkofer, J. prakt. Chem., 62 (1854), 508; M. Buchner, 
Ann., 96 (1885), 188-193. 

50. K. Strom, Arch. Pharm., 237 (1899), 525-543. 

51. J. A. Mjoen, Z. angew. Chem., 15 (1902), 97-111. 

52. M. X. Sullivan, J. Ind. Eng. Chem., 8 (1916), 1027. 

53. W. Gregory, Ann., 21 (1837), 143. 

54. H. B. Hill, Ber., 10 (1877), 936; 11 (1878), 456. 

55. D. Vorlander and K. Hobohm, Ber., 29 (1896), 1838. 



CHAPTER XIV 

THE FERMENTATION OF CELLULOSE AND WOOD BY 
BACTERIA AND FILAMENTOUS FUNGI 

The fungi have always played an important role in the life of 
man, but it is only within recent years that their economic possi- 
bilities have received the careful attention that they deserve. 
The utilization of wood waste of small dimension is still largely 
an unsolved problem. Nature has provided for the disposal of 
wood through the fungi. Controlled "decay" by organisms yield- 
ing products of value would appear to be a logical procedure, even 
though the trend of science has been for synthesis in vitro and 
the elimination of biological processes. 

Fermentation of Cellulose by Bacteria. — The bacteria able to 
decompose cellulose may be divided into two main groups, aerobic 
and anaerobic. In the former, fermentation takes place in the 
presence of free oxygen, and in the latter, in its absence. Prings- 
heim 1 gives the following classification : 

Group Fermentation Products 

1. Methane bacteria CH 4 , C0 2 , and lower fatty acids to butyric acid. 

2. Hydrogen bacteria H 2 , C0 2 , and lower fatty acids to butyric acid. 

3. Thermophilic bacteria. . CH 4 , H 2 , C0 2 , formic and acetic acids. 

4. Denitrifying bacteria . . . N 2 and C0 2 . 

In 1850, Mitscherlich 2 observed that the cell walls enclosing 
the starch granules in the potato were destroyed by microorgan- 
isms. He attributed the fermentation of the cellulose to vibrios 
which were present in large numbers. Popoff 3 made the first 
important contribution to the subject by showing the relation 
between the formation of methane in nature and the fermentation 
of cellulose. Filter paper fermented by slime from the sewers 

1 H. Pringsheim, Z. physiol. Chem., 78 (1912), 270. 

2 Mitscherlich, Monatsber. k. Acad. Wiss. Berlin (1850), 104; cited by 
F. Hoppe-Seyler, Z. physiol. Chem., 10 (1886), 403. 

3 L. Popoff, Arch. Physiol, 10 (1875), 113-146. 

470 



FERMENTATION OF CELLULOSE AND WOOD 471 

of Strassburg, 38 to 40° being the optimum temperature, gave a 
gas containing carbon dioxide, methane, hydrogen, and nitrogen. 
The gas collected after 4 weeks contained much less hydrogen 
than that collected at the end of 2 weeks. It was concluded 
that hydrogen formation was not due to the cellulose fermentation 
proper. 

The experiments of Hoppe-Seyler 1 supported the belief that 
the decomposition of cellulose was a methane fermentation. A 
flask containing a substrate of filter paper inoculated with sewage 
was kept for four years. The gas evolved contained carbon 
dioxide and methane but no hydrogen. Hoppe-Seyler concluded 
that the cellulose was first hydrolyzed to glucose, then the latter 
quantitatively converted into carbon dioxide and methane. 

C 6 H 12 6 = 3C0 2 + 3CH 4 . 

The living organism could not be distinguished from the Bacillus 
amylobacter described by van Tieghem 2 as fermenting certain 
natural celluloses. 

In 1895, Omelianski 3 contributed the first of a long series of 
papers on the fermentation of cellulose. A bacillus capable of 
fermenting filter paper was isolated from the mud of the river 
Niva. A similar organism, Bacillus fermentationis cellulosce, 4 
obtained from soil rich in vegetable debris, 5 gave under anaerobic 
conditions about 70 per cent of fatty acids, acetic, butyric, and 
valeric, and 30 per cent of hydrogen and methane. 

The cultures obtained from soil or horse manure were not pure, 6 
since according to experimental conditions a hydrogen or methane 
fermentation resulted. If the culture was first heated to 75° for 
about 15 minutes, and then held at 35°, the hydrogen fermentation 

1 F. Hoppe-Seyler, Z. physiol. Chem., 10 (1886), 201-217, 401-440; Ber., 
16 (1883), 122-123. 

2 P. E. L. van Tieghem, Bull. Soc. Botan. France, 24 (1877), 128-135; 
Compt. rend., 88 (1879), 205-210; Ibid., 89 (1879), 5-8. 

3 W. Omelianski, Compt. rend., 121 (1895), 363-365. 

4 W. Omelianski, Arch. Soc. Biolog. St. Petersb., 7 (1900), 411-434. 

5 W. Omelianski, Compt. rend., 125 (1877), 970-973, 1131-1133. 

6 W. Omelianski, Chem. Ztg., 26 (1902), 133; Centr. Bakt., Parasitenk., II 
Abt., 11 (1904), 369-377; cf. Ibid., II Abt., 8 (1902), 193-201, 225-231, 
257-263, 289-294, 321-326, 353-361, 385-391; II Abt., 12 (1904), 33-43; 
II Abt., 16 (1906), 673-687. 



472 CHEMISTRY OF CELLULOSE AND WOOD 

predominated, while without the preliminary heating the 
methane fermentation was much the stronger. Morphologi- 
cally, the two organisms were very similar. 

Kellerman and McBeth 1 isolated two species of cellulose- 
fermenting bacteria, and five species of contaminating bacteria, 
from Omelianski's hydrogen culture, and one cellulose-destroying 
and two contaminating species, from the methane culture. Con- 
trary to the results of Omelianski, the three cellulose-fermenting 
organisms isolated attacked the cellulose most rapidly under 
aerobic conditions, and did so without the production of gas. 

Five species of bacteria that ferment cellulose are described by 
McBeth and Scales. 2 The fermentation products were mainly 
the lower fatty acids. Gaseous products resulted from secondary 
fermentation by contaminating organisms, and not from the 
cellulose ferments. Not only the purity of the cultures of 
Kellerman and McBeth, but their ability to dissolve cellulose 
has been questioned. 3 

It has also been questioned whether a pure cellulose-fermenting 
organism has ever been isolated. There is evidence to the 
effect that cellulose is destroyed only by the symbiotic action 
of several organisms. Van Semis 4 found that cellulose was 
destroyed by the joint action of B. amylobacter and an organism 
isolated from the intestines of a rabbit, neither being effective 
alone. He concluded that methane was not formed directly. 
The cellulose was decomposed into hydrogen, carbon dioxide, 
and acetic acid, the hydrogen reducing the acetic acid to methane. 

The anaerobic decomposition of cellulose proceeds differently 
in the presence and absence of nitrates. 5 In their presence, the 

1 K. F. Kellerman and I. G. McBeth, Centr. Bakt, Parasitenk., II Abt., 
34 (1912), 63-64, 485-494; I. G. McBeth, Sail Science, 1 (1916), 437-487; 
cf. F. Lohnis and G. Lochhead, Centr. Bakt, Parasitenk., II Abt., 37 (1913), 
490-492. 

2 1. G. McBeth and F. M. Scales, "The Destruction of Cellulose by 
Bacteria and Filamentous Fungi/' Bull. 266, U. S. Dept. Agr. (1913). 

3 H. Pringsheim and S. Lichtenstein, Centr. Bakt., Parasitenk., II Abt., 
60 (1923), 309-311. 

4 A. H. C. van Senus, Jahresb. Fortschritte Lehre Garungs-Organismen, 1 
(1890), 136-139. 

5 C. van Iterson, Jun., Centr. Bakt, Parasitenk., II Abt., 11 (1904), 
689-698. 



FERMENTATION OF CELLULOSE AND WOOD 473 

cellulose is decomposed by denitrifying bacteria with the forma- 
tion of nitrogen, carbon dioxide, and water. Among aerobic, 
non-sporulating bacteria, Bacterium ferruginens is most impor- 
tant and active, in symbiosis with a yellow micrococcus itself 
inactive. 

Groenwege 1 studied the fermentation of cellulose, in a medium 
containing nitrates, using organisms from soil and septic tank 
liquid. There was denitrification and destruction of the cellu- 
lose, but as a result of the action of two classes of organisms. 
The a-, (3-, and 7-forms of Bacillus cellaresolvens attacked the 
cellulose, producing acetic, butyric, and lactic acids. These 
degradation products served as food for the denitrifying organ- 
isms, B. opalescens and B. viscosus, themselves incapable of 
attacking cellulose. 

Lohnis and Lochhead 2 consider the conclusion inevitable that 
the most active cellulose bacteria must have the symbiotic 
assistance of other microorganisms in the destruction of cellulose. 

A large number of organisms, whose individuality and power 
to ferment normal cellulose may be questioned, have been 
described. There may be mentioned: Bacillus cellulosoe desag- 
regans from the intestines of fowls; 3 Micrococcus cytophagus 
and M. melanocyclus from decayed radishes. 4 

An aerobic organism, (Spirochceta cytophaga) isolated from 
Rothhamsted soil by Hutchinson and Clayton, 5 decomposes 
cellulose readily with the formation of a yellow pigment, fatty 
acids, and a mucilage resembling pectin. Its optimum tempera- 
ture is 30°. Carbohydrates other than cellulose do not support 
growth. Advantage has been taken of this, and related organ- 
isms, to produce an artificial manure by the fermentation of 
straw to which nitrogenous substances have been added. 6 The 
fermentation, requiring about 3 months, is being commercialized 
as the "Adco" process. 

1 J. Groenwege, J. Soc. Chem. Ind., 40 (1921), 76A. 

2 F. Lohnis and G. Lochhead, Centr. Bakt., Parasitenk., II Abt, 68 (1923), 
434. 

3 A. Diastaso, Compt. rend. soc. biol., 70 (1912), 995. 

4 A. Peche, G. P. 292482 (1913). 

5 H. B. Hutchinson and J. Clayton, /. Agr. Sci., 9 (1919), 143-173. 

6 H. B. Hutchinson and E. H. Richards, J. Ministry Agr., 28 (1921), 
308-341; E. P. 152387 (1919); U. S. P. 1471979 (1923). 



474 CHEMISTRY OF CELLULOSE AND WOOD 

In purifying cultures of Spirochceta cytophaga, there was isolated 
a new type of aerobic, cellulose-fermenting organism, Microspira 
agar-liquefaciens, that also liquefies agar. 1 The organism appears 
to enter the fibers and attack the central canal. Certain small 
quantities of xylose and lignin stimulated fermentation of the 
cellulose (filter paper). "Straw gum" supported growth but 
not lignin, xylose, and arabinose. The results are scarcely 
indicative of the effect of constituents of the natural straw. The 
lignin must have been modified during isolation; and the straw 
gum that furnishes xylose did not have a stimulating action. 

The thermophilic bacteria are of great interest on account of 
the speed with which cellulose is decomposed. In 1889, Mac- 
fadyen and Blaxall 2 recognized the group of thermophilic bac- 
teria. Viscose was almost completely fermented, at 60°, in 3 
weeks, acetic and butyric acids being formed. Failure to 
ferment pure cellulose with this group may sometimes be due 
to the absence of iron. Kronlik 3 found that a trace of ferric 
chloride produced an intense decomposition. 

Butyric acid does not appear to be formed by thermophilic 
bacteria. At 55 to 60°, anaerobic bacteria gave only hydrogen, 
carbon dioxide, formic acid, and acetic acid, the latter greatly 
predominating. Fermentation of 3 grams of cellulose gave 
0.2125 gram of formic acid and 1.152 grams of acetic acid. The